diff --git a/.github/scripts/eval_regression_api.py b/.github/scripts/eval_regression_api.py
index ba1902a9..98f0fdf0 100644
--- a/.github/scripts/eval_regression_api.py
+++ b/.github/scripts/eval_regression_api.py
@@ -24,9 +24,9 @@ models = [
         abbr='lmdeploy-api-test',
         type=OpenAISDK,
         key='EMPTY',
-        openai_api_base='http://0.0.0.0:23333/v1',
-        path='internlm2',
-        tokenizer_path='internlm/internlm2_5-7b-chat',
+        openai_api_base='http://localhost:23333/v1',
+        path='internlm3',
+        tokenizer_path='internlm/internlm3-8b-instruct',
         rpm_verbose=True,
         meta_template=api_meta_template,
         query_per_second=128,
diff --git a/.github/scripts/eval_regression_base_models.py b/.github/scripts/eval_regression_base_models.py
index a8dc7a60..4259cc36 100644
--- a/.github/scripts/eval_regression_base_models.py
+++ b/.github/scripts/eval_regression_base_models.py
@@ -11,18 +11,10 @@ with read_base():
     from opencompass.configs.datasets.winogrande.winogrande_5shot_ll_252f01 import \
         winogrande_datasets  # noqa: F401, E501
     # read hf models - chat models
-    from opencompass.configs.models.chatglm.hf_glm4_9b import \
-        models as hf_glm4_9b_model  # noqa: F401, E501
     from opencompass.configs.models.chatglm.lmdeploy_glm4_9b import \
         models as lmdeploy_glm4_9b_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.hf_deepseek_7b_base import \
         models as hf_deepseek_7b_base_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_67b_base import \
-        models as hf_deepseek_67b_base_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_moe_16b_base import \
-        models as hf_deepseek_moe_16b_base_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_v2_lite import \
-        models as hf_deepseek_v2_lite_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.lmdeploy_deepseek_7b_base import \
         models as lmdeploy_deepseek_7b_base_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.lmdeploy_deepseek_67b_base import \
@@ -49,12 +41,6 @@ with read_base():
         models as hf_internlm2_5_7b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.hf_internlm2_7b import \
         models as hf_internlm2_7b_model  # noqa: F401, E501
-    from opencompass.configs.models.hf_internlm.hf_internlm2_20b import \
-        models as hf_internlm2_20b_model  # noqa: F401, E501
-    from opencompass.configs.models.hf_internlm.hf_internlm2_base_7b import \
-        models as hf_internlm2_base_7b_model  # noqa: F401, E501
-    from opencompass.configs.models.hf_internlm.hf_internlm2_base_20b import \
-        models as hf_internlm2_base_20b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_1_8b import \
         models as lmdeploy_internlm2_1_8b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_5_7b import \
@@ -65,14 +51,14 @@ with read_base():
         models as lmdeploy_internlm2_20b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_base_7b import \
         models as lmdeploy_internlm2_base_7b_model  # noqa: F401, E501
+    from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_base_20b import \
+        models as lmdeploy_internlm2_base_20b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.hf_llama2_7b import \
         models as hf_llama2_7b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.hf_llama3_1_8b import \
         models as hf_llama3_1_8b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.hf_llama3_8b import \
         models as hf_llama3_8b_model  # noqa: F401, E501
-    from opencompass.configs.models.hf_llama.hf_llama3_70b import \
-        models as hf_llama3_70b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.lmdeploy_llama3_1_8b import \
         models as lmdeploy_llama3_1_8b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.lmdeploy_llama3_8b import \
diff --git a/.github/scripts/eval_regression_chat_models.py b/.github/scripts/eval_regression_chat_models.py
index 40ec1bc5..bfe923f6 100644
--- a/.github/scripts/eval_regression_chat_models.py
+++ b/.github/scripts/eval_regression_chat_models.py
@@ -15,14 +15,24 @@ with read_base():
         models as vllm_glm4_9b_chat_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.hf_deepseek_7b_chat import \
         models as hf_deepseek_7b_chat_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_67b_chat import \
-        models as hf_deepseek_67b_chat_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_moe_16b_chat import \
-        models as hf_deepseek_moe_16b_chat_model  # noqa: F401, E501
-    from opencompass.configs.models.deepseek.hf_deepseek_v2_lite_chat import \
-        models as hf_deepseek_v2_lite_chat_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_67b_chat import \
+        models as lmdeploy_deepseek_67b_chat_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_r1_distill_llama_8b import \
+        models as \
+        lmdeploy_deepseek_r1_distill_llama_8b_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_r1_distill_llama_70b import \
+        models as \
+        lmdeploy_deepseek_r1_distill_llama_70b_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_r1_distill_qwen_1_5b import \
+        models as \
+        lmdeploy_deepseek_r1_distill_qwen_1_5b_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_r1_distill_qwen_32b import \
+        models as \
+        lmdeploy_deepseek_r1_distill_qwen_32b_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.lmdeploy_deepseek_v2_5_1210 import \
         models as lmdeploy_deepseek_v2_5_1210_model  # noqa: F401, E501
+    from opencompass.configs.models.deepseek.lmdeploy_deepseek_v2_lite import \
+        models as lmdeploy_deepseek_v2_lite_model  # noqa: F401, E501
     from opencompass.configs.models.deepseek.vllm_deepseek_7b_chat import \
         models as vllm_deepseek_7b_chat_model  # noqa: F401, E501
     from opencompass.configs.models.gemma.hf_gemma2_2b_it import \
@@ -45,6 +55,8 @@ with read_base():
         models as hf_internlm2_5_7b_chat_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.hf_internlm2_5_20b_chat import \
         models as hf_internlm2_5_20b_chat_model  # noqa: F401, E501
+    from opencompass.configs.models.hf_internlm.hf_internlm3_8b_instruct import \
+        models as hf_internlm3_8b_instruct_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_5_7b_chat import \
         models as lmdeploy_internlm2_5_7b_chat_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_5_20b_chat import \
@@ -57,6 +69,8 @@ with read_base():
         models as lmdeploy_internlm2_chat_7b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.lmdeploy_internlm2_chat_7b_sft import \
         models as lmdeploy_internlm2_chat_7b_sft_model  # noqa: F401, E501
+    from opencompass.configs.models.hf_internlm.lmdeploy_internlm3_8b_instruct import \
+        models as lmdeploy_internlm3_8b_instruct_model  # noqa: F401, E501
     from opencompass.configs.models.hf_internlm.vllm_internlm2_chat_7b import \
         models as vllm_internlm2_chat_7b_model  # noqa: F401, E501
     from opencompass.configs.models.hf_llama.hf_llama3_1_8b_instruct import \
@@ -83,10 +97,6 @@ with read_base():
         models as hf_mistral_nemo_instruct_2407_model  # noqa: F401, E501
     from opencompass.configs.models.mistral.hf_mistral_small_instruct_2409 import \
         models as hf_mistral_small_instruct_2409_model  # noqa: F401, E501
-    from opencompass.configs.models.mistral.hf_mixtral_8x7b_instruct_v0_1 import \
-        models as hf_mixtral_8x7b_instruct_v0_1_model  # noqa: F401, E501
-    from opencompass.configs.models.mistral.hf_mixtral_8x22b_instruct_v0_1 import \
-        models as hf_mixtral_8x22b_instruct_v0_1_model  # noqa: F401, E501
     from opencompass.configs.models.mistral.lmdeploy_mistral_large_instruct_2411 import \
         models as \
         lmdeploy_mistral_large_instruct_2411_model  # noqa: F401, E501
@@ -95,14 +105,19 @@ with read_base():
     from opencompass.configs.models.mistral.lmdeploy_mistral_small_instruct_2409 import \
         models as \
         lmdeploy_mistral_small_instruct_2409_model  # noqa: F401, E501
+    from opencompass.configs.models.mistral.lmdeploy_mixtral_8x22b_instruct_v0_1 import \
+        models as \
+        lmdeploy_mixtral_8x22b_instruct_v0_1_model  # noqa: F401, E501
     from opencompass.configs.models.mistral.vllm_mistral_7b_instruct_v0_1 import \
         models as vllm_mistral_7b_instruct_v0_1_model  # noqa: F401, E501
     from opencompass.configs.models.mistral.vllm_mistral_7b_instruct_v0_2 import \
         models as vllm_mistral_7b_instruct_v0_2_model  # noqa: F401, E501
+    from opencompass.configs.models.mistral.vllm_mixtral_8x22b_instruct_v0_1 import \
+        models as vllm_mixtral_8x22b_instruct_v0_1_model  # noqa: F401, E501
     from opencompass.configs.models.nvidia.lmdeploy_nemotron_70b_instruct_hf import \
         models as lmdeploy_nemotron_70b_instruct_hf_model  # noqa: F401, E501
-    from opencompass.configs.models.phi.hf_phi_3_mini_4k_instruct import \
-        models as hf_phi_3_mini_4k_instruct_model  # noqa: F401, E501
+    from opencompass.configs.models.phi.hf_phi_4 import \
+        models as hf_phi_4_model  # noqa: F401, E501
     from opencompass.configs.models.qwen2_5.hf_qwen2_5_0_5b_instruct import \
         models as hf_qwen2_5_0_5b_instruct_model  # noqa: F401, E501
     from opencompass.configs.models.qwen2_5.hf_qwen2_5_3b_instruct import \
@@ -142,6 +157,8 @@ with read_base():
 
     from ...volc import infer as volc_infer  # noqa: F401, E501
 
+hf_glm4_9b_chat_model[0]['path'] = 'THUDM/glm-4-9b-chat-hf'
+
 race_datasets = [race_datasets[1]]
 datasets = sum([v for k, v in locals().items() if k.endswith('_datasets')], [])
 
diff --git a/.github/scripts/oc_score_assert.py b/.github/scripts/oc_score_assert.py
index 4ef414dc..1cbc5ad2 100644
--- a/.github/scripts/oc_score_assert.py
+++ b/.github/scripts/oc_score_assert.py
@@ -175,10 +175,11 @@ class TestApibench:
 class TestVolcFullbench:
     """Test cases for chat model."""
 
-    @pytest.mark.parametrize(
-        'model, dataset',
-        [(p1, p2) for p1 in ['internlm2_5-7b-chat-turbomind']
-         for p2 in dataset_list('internlm2_5-7b-chat-turbomind', 'objective')])
+    @pytest.mark.parametrize('model, dataset', [(p1, p2) for p1 in [
+        'internlm2_5-7b-chat-turbomind', 'qwen2.5-7b-instruct-turbomind',
+        'internlm2_5-7b-chat-pytorch', 'qwen2.5-7b-instruct-pytorch',
+        'internlm3-8b-instruct-turbomind', 'internlm3-8b-instruct-pytorch'
+    ] for p2 in dataset_list(p1, 'objective')])
     @pytest.mark.chat_objective
     def test_chat_objective(self, baseline_scores_fullbench, result_scores,
                             model, dataset):
@@ -245,10 +246,7 @@ class TestCmdCase:
     @pytest.mark.parametrize('model, dataset',
                              [('internlm2_5-7b-hf', 'race-middle_accuracy'),
                               ('internlm2_5-7b-hf', 'race-high_accuracy'),
-                              ('internlm2_5-7b-hf', 'demo_gsm8k_accuracy'),
-                              ('internlm2-1.8b-hf', 'race-middle_accuracy'),
-                              ('internlm2-1.8b-hf', 'race-high_accuracy'),
-                              ('internlm2-1.8b-hf', 'demo_gsm8k_accuracy')])
+                              ('internlm2_5-7b-hf', 'demo_gsm8k_accuracy')])
     def test_cmd_case1(self, baseline_scores, result_scores, model, dataset):
         base_score = baseline_scores.get(model).get(dataset)
         result_score = result_scores.get(model).get(dataset)
@@ -260,9 +258,9 @@ class TestCmdCase:
         [('internlm2_5-7b-chat-lmdeploy', 'race-middle_accuracy'),
          ('internlm2_5-7b-chat-lmdeploy', 'race-high_accuracy'),
          ('internlm2_5-7b-chat-lmdeploy', 'demo_gsm8k_accuracy'),
-         ('internlm2-chat-1.8b-lmdeploy', 'race-middle_accuracy'),
-         ('internlm2-chat-1.8b-lmdeploy', 'race-high_accuracy'),
-         ('internlm2-chat-1.8b-lmdeploy', 'demo_gsm8k_accuracy')])
+         ('internlm3-8b-instruct-lmdeploy', 'race-middle_accuracy'),
+         ('internlm3-8b-instruct-lmdeploy', 'race-high_accuracy'),
+         ('internlm3-8b-instruct-lmdeploy', 'demo_gsm8k_accuracy')])
     def test_cmd_case2(self, baseline_scores, result_scores, model, dataset):
         base_score = baseline_scores.get(model).get(dataset)
         result_score = result_scores.get(model).get(dataset)
@@ -280,13 +278,25 @@ class TestCmdCase:
 
     @pytest.mark.case4
     @pytest.mark.parametrize(
-        'model, dataset', [('internlm2_5-7b-chat_hf', 'race-middle_accuracy'),
-                           ('internlm2_5-7b-chat_hf', 'race-high_accuracy'),
-                           ('internlm2_5-7b-chat_hf', 'demo_gsm8k_accuracy')])
+        'model, dataset',
+        [('internlm3-8b-instruct_hf-lmdeploy', 'race-middle_accuracy'),
+         ('internlm3-8b-instruct_hf-lmdeploy', 'race-high_accuracy'),
+         ('internlm3-8b-instruct_hf-lmdeploy', 'demo_gsm8k_accuracy')])
     def test_cmd_case4(self, baseline_scores, result_scores, model, dataset):
         base_score = baseline_scores.get(model).get(dataset)
         result_score = result_scores.get(model).get(dataset)
-        assert_score(model, result_score, base_score, dataset)
+        assert_score(model + '_batch', result_score, base_score, dataset)
+
+    @pytest.mark.case5
+    @pytest.mark.parametrize(
+        'model, dataset',
+        [('internlm3-8b-instruct_hf-vllm', 'race-middle_accuracy'),
+         ('internlm3-8b-instruct_hf-vllm', 'race-high_accuracy'),
+         ('internlm3-8b-instruct_hf-vllm', 'demo_gsm8k_accuracy')])
+    def test_cmd_case5(self, baseline_scores, result_scores, model, dataset):
+        base_score = baseline_scores.get(model).get(dataset)
+        result_score = result_scores.get(model).get(dataset)
+        assert_score(model + '_batch', result_score, base_score, dataset)
 
 
 def assert_score(model_type, score, baseline, dataset: str = ''):
diff --git a/.github/scripts/oc_score_baseline.yaml b/.github/scripts/oc_score_baseline.yaml
index cd2e3328..e4567553 100644
--- a/.github/scripts/oc_score_baseline.yaml
+++ b/.github/scripts/oc_score_baseline.yaml
@@ -8,20 +8,25 @@ internlm2_5-7b_hf:
     race-middle_accuracy: 91.78
     race-high_accuracy: 90.02
 
-internlm2-1.8b-hf:
-    demo_gsm8k_accuracy: 15.62
-    race-middle_accuracy: 71.66
-    race-high_accuracy: 66.38
-
 internlm2_5-7b-chat-lmdeploy:
-    demo_gsm8k_accuracy: 89.06
+    demo_gsm8k_accuracy: 87.50
     race-middle_accuracy: 92.76
     race-high_accuracy: 90.54
 
-internlm2-chat-1.8b-lmdeploy:
-    demo_gsm8k_accuracy: 31
-    race-middle_accuracy: 81.34
-    race-high_accuracy: 73.96
+internlm3-8b-instruct-lmdeploy:
+    demo_gsm8k_accuracy: 73.44
+    race-middle_accuracy: 93.38
+    race-high_accuracy: 90.34
+
+internlm3-8b-instruct_hf-lmdeploy:
+    demo_gsm8k_accuracy: 73.44
+    race-middle_accuracy: 93.38
+    race-high_accuracy: 90.34
+
+internlm3-8b-instruct_hf-vllm:
+    demo_gsm8k_accuracy: 81.25
+    race-middle_accuracy: 92.20
+    race-high_accuracy: 89.88
 
 internlm2_5-7b-chat_hf:
     demo_gsm8k_accuracy: 87.50
@@ -29,6 +34,6 @@ internlm2_5-7b-chat_hf:
     race-high_accuracy: 90.48
 
 lmdeploy-api-test:
-    gsm8k_accuracy: 68.75
-    race-middle_accuracy: 87.50
+    gsm8k_accuracy: 56.25
+    race-middle_accuracy: 93.75
     race-high_accuracy: 93.75
diff --git a/.github/scripts/oc_score_baseline_fullbench.yaml b/.github/scripts/oc_score_baseline_fullbench.yaml
index c0e735fb..3f5753d3 100644
--- a/.github/scripts/oc_score_baseline_fullbench.yaml
+++ b/.github/scripts/oc_score_baseline_fullbench.yaml
@@ -39,15 +39,15 @@ internlm2_5-7b-chat-hf_fullbench:
         college_knowledge_naive_average: 87.5
     subjective:
         alignment_bench_v1_1_总分: 0.66
-        alpaca_eval_total: 20
+        alpaca_eval_total: 0
         arenahard_score: 50
         Followbench_naive_average: 1
         CompassArena_naive_average: 43
         mtbench101_avg: 7.8
-        wildbench_average: -12.78
+        wildbench_average: -15.56
         simpleqa_accuracy_given_attempted: 0
         chinese_simpleqa_given_attempted_accuracy: 1
-        alignment_bench_v1_1_专业能力: 7.90
+        alignment_bench_v1_1_专业能力: 8.00
         alignment_bench_v1_1_数学计算: 0
         alignment_bench_v1_1_基本任务: 0
         alignment_bench_v1_1_逻辑推理: 0
@@ -55,7 +55,7 @@ internlm2_5-7b-chat-hf_fullbench:
         alignment_bench_v1_1_文本写作: 0
         alignment_bench_v1_1_角色扮演: 0
         alignment_bench_v1_1_综合问答: 0
-        alpaca_eval_helpful_base: 20
+        alpaca_eval_helpful_base: 0
         compassarena_language_naive_average: 35
         compassarena_knowledge_naive_average: 55
         compassarena_reason_v2_naive_average: 40
@@ -78,53 +78,53 @@ internlm2_5-7b-chat-hf_fullbench:
 internlm2_5-7b-chat-turbomind_fullbench:
     objective:
         race-high_accuracy:  93.75
-        ARC-c_accuracy: 93.75
+        ARC-c_accuracy: 87.50
         BoolQ_accuracy: 68.75
         triviaqa_wiki_1shot_score: 50
         nq_open_1shot_score: 25
         IFEval_Prompt-level-strict-accuracy: 56.25
-        drop_accuracy: 81.25
+        drop_accuracy: 75
         GPQA_diamond_accuracy: 31.25
-        hellaswag_accuracy: 81.25
-        TheoremQA_score: 6.25
+        hellaswag_accuracy: 87.5
+        TheoremQA_score: 12.5
         musr_average_naive_average: 39.58
-        korbench_single_naive_average: 37.50
-        gsm8k_accuracy: 68.75
-        math_accuracy: 68.75
+        korbench_single_naive_average: 40
+        gsm8k_accuracy: 62.5
+        math_accuracy: 75
         cmo_fib_accuracy: 6.25
         aime2024_accuracy: 6.25
-        wikibench-wiki-single_choice_cncircular_perf_4: 50.00
+        wikibench-wiki-single_choice_cncircular_perf_4: 25
         sanitized_mbpp_score: 68.75
-        ds1000_naive_average: 16.96
+        ds1000_naive_average: 17.86
         lcb_code_generation_pass@1: 12.5
         lcb_code_execution_pass@1: 43.75
-        lcb_test_output_pass@1: 25.00
-        bbh-logical_deduction_seven_objects_score: 50.00
-        bbh-multistep_arithmetic_two_score: 68.75
-        mmlu-other_accuracy: 69.71
-        cmmlu-china-specific_accuracy: 75.83
+        lcb_test_output_pass@1: 18.75
+        bbh-logical_deduction_seven_objects_score: 56.25
+        bbh-multistep_arithmetic_two_score: 75
+        mmlu-other_accuracy: 72.6
+        cmmlu-china-specific_accuracy: 78.33
         mmlu_pro_math_accuracy: 31.25
-        ds1000_Pandas_accuracy: 0
+        ds1000_Pandas_accuracy: 12.5
         ds1000_Numpy_accuracy: 0
         ds1000_Tensorflow_accuracy: 12.5
-        ds1000_Scipy_accuracy: 18.75
+        ds1000_Scipy_accuracy: 25
         ds1000_Sklearn_accuracy: 18.75
-        ds1000_Pytorch_accuracy: 18.75
+        ds1000_Pytorch_accuracy: 6.25
         ds1000_Matplotlib_accuracy: 50.00
         openai_mmmlu_lite_AR-XY_accuracy: 37.5
         college_naive_average: 12.50
         college_knowledge_naive_average: 87.5
     subjective:
-        alignment_bench_v1_1_总分: 0.70
+        alignment_bench_v1_1_总分: 0.66
         alpaca_eval_total: 0
         arenahard_score: 50
         Followbench_naive_average: 1
-        CompassArena_naive_average: 38
-        mtbench101_avg: 7.80
-        wildbench_average: -4.86
+        CompassArena_naive_average: 40
+        mtbench101_avg: 8
+        wildbench_average: -6.81
         simpleqa_accuracy_given_attempted: 0
         chinese_simpleqa_given_attempted_accuracy: 1
-        alignment_bench_v1_1_专业能力: 8.4
+        alignment_bench_v1_1_专业能力: 7.9
         alignment_bench_v1_1_数学计算: 0
         alignment_bench_v1_1_基本任务: 0
         alignment_bench_v1_1_逻辑推理: 0
@@ -134,10 +134,10 @@ internlm2_5-7b-chat-turbomind_fullbench:
         alignment_bench_v1_1_综合问答: 0
         alpaca_eval_helpful_base: 0
         compassarena_language_naive_average: 35
-        compassarena_knowledge_naive_average: 50
-        compassarena_reason_v2_naive_average: 30
-        compassarena_math_v2_naive_average: 50
-        compassarena_creationv2_zh_naive_average: 25
+        compassarena_knowledge_naive_average: 45
+        compassarena_reason_v2_naive_average: 25
+        compassarena_math_v2_naive_average: 60
+        compassarena_creationv2_zh_naive_average: 35
         followbench_llmeval_en_HSR_AVG: 1
         followbench_llmeval_en_SSR_AVG: 1
         followbench_llmeval_en_HSR_L1: 1
@@ -190,20 +190,20 @@ internlm2_5-7b-turbomind_fullbench:
         drop_accuracy: 62.5
         GPQA_diamond_accuracy: 62.5
         hellaswag_accuracy: 93.75
-        TheoremQA_score: 25.00
+        TheoremQA_score: 31.25
         winogrande_accuracy: 87.5
-        gsm8k_accuracy: 62.50
-        GaokaoBench_2010-2022_Math_II_MCQs_score: 81.25
+        gsm8k_accuracy: 56.25
+        GaokaoBench_2010-2022_Math_II_MCQs_score: 68.75
         GaokaoBench_2010-2022_Math_II_Fill-in-the-Blank_score: 0
         math_accuracy: 18.75
         wikibench-wiki-single_choice_cncircular_perf_4: 25
         sanitized_mbpp_score: 62.50
-        dingo_en_192_score: 31.25
+        dingo_en_192_score: 50.00
         dingo_zh_170_score: 93.75
         mmlu-other_accuracy: 76.92
         cmmlu-china-specific_accuracy: 84.17
         mmlu_pro_math_accuracy: 18.75
-        bbh-logical_deduction_seven_objects_score: 50
+        bbh-logical_deduction_seven_objects_score: 43.75
         bbh-multistep_arithmetic_two_score: 56.25
         college_naive_average: 12.5
         college_knowledge_naive_average: 87.5
@@ -409,7 +409,7 @@ internlm2_5-7b-chat-turbomind:
         alpaca_eval_koala: 28.21
         alpaca_eval_oasst: 23.4
         alpaca_eval_selfinstruct: 30.95
-        alpaca_eval_vicuna: 25
+        alpaca_eval_vicuna: 33.75
         compassarena_language_naive_average: 52.5
         compassarena_knowledge_naive_average: 36
         compassarena_reason_v2_naive_average: 35
@@ -454,3 +454,530 @@ internlm2_5-7b-chat-1m-turbomind:
         longbench_few-shot-learning_score: 51.67
         longbench_synthetic-tasks_score: 66.83
         longbench_code-completion_score: 45.99
+
+
+qwen2.5-7b-instruct-turbomind:
+    objective:
+        race-high_accuracy: 84.99
+        ARC-c_accuracy: 92.2
+        BoolQ_accuracy: 86.7
+        triviaqa_wiki_1shot_score: 53.06
+        nq_open_1shot_score: 17.51
+        mmmlu_lite_naive_average: 54.96
+        IFEval_Prompt-level-strict-accuracy: 71.53
+        drop_accuracy: 80.07
+        bbh_naive_average: 68.81
+        GPQA_diamond_accuracy: 34.34
+        hellaswag_accuracy: 85.42
+        TheoremQA_score: 18.38
+        musr_average_naive_average: 43.44
+        korbench_single_naive_average: 39.44
+        ARC_Prize_Public_Evaluation_accuracy: 0
+        gsm8k_accuracy: 92.57
+        GaokaoBench_weighted_average: 80.14
+        math_accuracy: 73.58
+        cmo_fib_accuracy: 25
+        aime2024_accuracy: 16.67
+        Mathbench_naive_average: 77.33
+        wikibench-wiki-single_choice_cncircular_perf_4: 34.9
+        cmmlu_naive_average: 75.97
+        mmlu_naive_average: 76.01
+        mmlu_pro_naive_average: 56.12
+        openai_humaneval_humaneval_pass@1: 83.54
+        sanitized_mbpp_score: 74.71
+        humanevalx_naive_average: 48.29
+        ds1000_naive_average: 18.66
+        lcb_code_generation_pass@1: 39.5
+        lcb_code_execution_pass@1: 42.38
+        lcb_test_output_pass@1: 50.68
+        bigcodebench_hard_instruct_pass@1: 16.22
+        bigcodebench_hard_complete_pass@1: 11.49
+        teval_naive_average: 79.72
+        SciCode_sub_accuracy: 100
+        qa_dingo_cn_score: 99.01
+        mmlu_accuracy: 76.01
+        mmlu-stem_accuracy: 77.59
+        mmlu-social-science_accuracy: 79.02
+        mmlu-humanities_accuracy: 72.07
+        mmlu-other_accuracy: 74.86
+        cmmlu_accuracy: 75.97
+        cmmlu-stem_accuracy: 73.09
+        cmmlu-social-science_accuracy: 75.95
+        cmmlu-humanities_accuracy: 76.53
+        cmmlu-other_accuracy: 78.79
+        cmmlu-china-specific_accuracy: 73.17
+        mmlu_pro_accuracy: 56.12
+        mmlu_pro_biology_accuracy: 71.41
+        mmlu_pro_business_accuracy: 67.68
+        mmlu_pro_chemistry_accuracy: 54.59
+        mmlu_pro_computer_science_accuracy: 58.29
+        mmlu_pro_economics_accuracy: 66.82
+        mmlu_pro_engineering_accuracy: 42.41
+        mmlu_pro_health_accuracy: 55.87
+        mmlu_pro_history_accuracy: 46.46
+        mmlu_pro_law_accuracy: 28.97
+        mmlu_pro_math_accuracy: 73.13
+        mmlu_pro_philosophy_accuracy: 44.89
+        mmlu_pro_physics_accuracy: 58.43
+        mmlu_pro_psychology_accuracy: 63.16
+        mmlu_pro_other_accuracy: 53.57
+        humanevalx-python_pass@1: 50
+        humanevalx-cpp_pass@1: 42.07
+        humanevalx-go_pass@1: 0
+        humanevalx-java_pass@1: 74.39
+        humanevalx-js_pass@1: 75
+        ds1000_Pandas_accuracy: 14.09
+        ds1000_Numpy_accuracy: 8.18
+        ds1000_Tensorflow_accuracy: 17.78
+        ds1000_Scipy_accuracy: 15.09
+        ds1000_Sklearn_accuracy: 10.43
+        ds1000_Pytorch_accuracy: 4.41
+        ds1000_Matplotlib_accuracy: 60.65
+        mmmlu_lite_accuracy: 54.96
+        openai_mmmlu_lite_AR-XY_accuracy: 42.32
+        openai_mmmlu_lite_BN-BD_accuracy: 42.25
+        openai_mmmlu_lite_DE-DE_accuracy: 59.93
+        openai_mmmlu_lite_ES-LA_accuracy: 66.53
+        openai_mmmlu_lite_FR-FR_accuracy: 66.88
+        openai_mmmlu_lite_HI-IN_accuracy: 49.26
+        openai_mmmlu_lite_ID-ID_accuracy: 61.26
+        openai_mmmlu_lite_IT-IT_accuracy: 65.47
+        openai_mmmlu_lite_JA-JP_accuracy: 61.54
+        openai_mmmlu_lite_KO-KR_accuracy: 60.28
+        openai_mmmlu_lite_PT-BR_accuracy: 55.51
+        openai_mmmlu_lite_SW-KE_accuracy: 36.42
+        openai_mmmlu_lite_YO-NG_accuracy: 32.14
+        openai_mmmlu_lite_ZH-CN_accuracy: 69.61
+        college_naive_average: 48
+        high_naive_average: 59
+        middle_naive_average: 78
+        primary_naive_average: 85.67
+        arithmetic_naive_average: 75.67
+        mathbench-a (average)_naive_average: 69.27
+        college_knowledge_naive_average: 83.86
+        high_knowledge_naive_average: 80.29
+        middle_knowledge_naive_average: 84.26
+        primary_knowledge_naive_average: 93.16
+        mathbench-t (average)_naive_average: 85.39
+
+
+
+
+internlm2_5-7b-chat-pytorch:
+    objective:
+        race-high_accuracy: 86.39
+        ARC-c_accuracy: 90.51
+        BoolQ_accuracy: 88.01
+        triviaqa_wiki_1shot_score: 64.77
+        nq_open_1shot_score: 22.71
+        mmmlu_lite_naive_average: 45.02
+        IFEval_Prompt-level-strict-accuracy: 56.56
+        drop_accuracy: 75.46
+        bbh_naive_average: 73.34
+        GPQA_diamond_accuracy: 32.83
+        hellaswag_accuracy: 94.81
+        TheoremQA_score: 23.88
+        musr_average_naive_average: 51.31
+        korbench_single_naive_average: 32
+        ARC_Prize_Public_Evaluation_accuracy: 0.01
+        gsm8k_accuracy: 86.96
+        GaokaoBench_weighted_average: 78.05
+        math_accuracy: 60.34
+        cmo_fib_accuracy: 12.98
+        aime2024_accuracy: 3.33
+        Mathbench_naive_average: 64.82
+        wikibench-wiki-single_choice_cncircular_perf_4: 31.7
+        cmmlu_naive_average: 74.24
+        mmlu_naive_average: 70.2
+        mmlu_pro_naive_average: 45.39
+        openai_humaneval_humaneval_pass@1: 70.12
+        sanitized_mbpp_score: 64.59
+        humanevalx_naive_average: 38.78
+        ds1000_naive_average: 14.19
+        lcb_code_generation_pass@1: 16.5
+        lcb_code_execution_pass@1: 33.82
+        lcb_test_output_pass@1: 22.62
+        bigcodebench_hard_instruct_pass@1: 6.08
+        bigcodebench_hard_complete_pass@1: 6.76
+        teval_naive_average: 79.73
+        SciCode_sub_accuracy: 100
+        qa_dingo_cn_score: 100
+        mmlu_accuracy: 70.2
+        mmlu-stem_accuracy: 67.73
+        mmlu-social-science_accuracy: 75.49
+        mmlu-humanities_accuracy: 68.56
+        mmlu-other_accuracy: 70.58
+        cmmlu_accuracy: 74.24
+        cmmlu-stem_accuracy: 66.7
+        cmmlu-social-science_accuracy: 75.88
+        cmmlu-humanities_accuracy: 77.56
+        cmmlu-other_accuracy: 77.52
+        cmmlu-china-specific_accuracy: 73.46
+        mmlu_pro_accuracy: 45.39
+        mmlu_pro_biology_accuracy: 65.83
+        mmlu_pro_business_accuracy: 51.96
+        mmlu_pro_chemistry_accuracy: 36.84
+        mmlu_pro_computer_science_accuracy: 48.29
+        mmlu_pro_economics_accuracy: 56.16
+        mmlu_pro_engineering_accuracy: 29.1
+        mmlu_pro_health_accuracy: 44.5
+        mmlu_pro_history_accuracy: 42.26
+        mmlu_pro_law_accuracy: 24.98
+        mmlu_pro_math_accuracy: 54.85
+        mmlu_pro_philosophy_accuracy: 39.28
+        mmlu_pro_physics_accuracy: 37.41
+        mmlu_pro_psychology_accuracy: 58.27
+        mmlu_pro_other_accuracy: 45.78
+        humanevalx-python_pass@1: 56.1
+        humanevalx-cpp_pass@1: 20.73
+        humanevalx-go_pass@1: 0
+        humanevalx-java_pass@1: 59.15
+        humanevalx-js_pass@1: 57.93
+        ds1000_Pandas_accuracy: 8.93
+        ds1000_Numpy_accuracy: 4.09
+        ds1000_Tensorflow_accuracy: 11.11
+        ds1000_Scipy_accuracy: 7.55
+        ds1000_Sklearn_accuracy: 7.83
+        ds1000_Pytorch_accuracy: 8.82
+        ds1000_Matplotlib_accuracy: 50.97
+        mmmlu_lite_accuracy: 45.02
+        openai_mmmlu_lite_AR-XY_accuracy: 18.6
+        openai_mmmlu_lite_BN-BD_accuracy: 27.58
+        openai_mmmlu_lite_DE-DE_accuracy: 51.23
+        openai_mmmlu_lite_ES-LA_accuracy: 56.63
+        openai_mmmlu_lite_FR-FR_accuracy: 58.11
+        openai_mmmlu_lite_HI-IN_accuracy: 33.82
+        openai_mmmlu_lite_ID-ID_accuracy: 50.39
+        openai_mmmlu_lite_IT-IT_accuracy: 50.39
+        openai_mmmlu_lite_JA-JP_accuracy: 50.95
+        openai_mmmlu_lite_KO-KR_accuracy: 45.05
+        openai_mmmlu_lite_PT-BR_accuracy: 57.89
+        openai_mmmlu_lite_SW-KE_accuracy: 32.14
+        openai_mmmlu_lite_YO-NG_accuracy: 32.14
+        openai_mmmlu_lite_ZH-CN_accuracy: 65.33
+        college_naive_average: 21
+        high_naive_average: 47
+        middle_naive_average: 59.67
+        primary_naive_average: 76
+        arithmetic_naive_average: 62
+        mathbench-a (average)_naive_average: 53.13
+        college_knowledge_naive_average: 68.99
+        high_knowledge_naive_average: 70.06
+        middle_knowledge_naive_average: 78.53
+        primary_knowledge_naive_average: 88.49
+        mathbench-t (average)_naive_average: 76.51
+
+
+qwen2.5-7b-instruct-pytorch:
+    objective:
+        race-high_accuracy: 85.16
+        ARC-c_accuracy: 90.85
+        BoolQ_accuracy: 86.61
+        triviaqa_wiki_1shot_score: 52.96
+        nq_open_1shot_score: 17.62
+        mmmlu_lite_naive_average: 54.7
+        IFEval_Prompt-level-strict-accuracy: 71.35
+        drop_accuracy: 80.23
+        bbh_naive_average: 68.88
+        GPQA_diamond_accuracy: 36.36
+        hellaswag_accuracy: 85.49
+        TheoremQA_score: 18.38
+        musr_average_naive_average: 43.3
+        korbench_single_naive_average: 39.44
+        ARC_Prize_Public_Evaluation_accuracy: 0
+        gsm8k_accuracy: 91.66
+        GaokaoBench_weighted_average: 80.02
+        math_accuracy: 73.74
+        cmo_fib_accuracy: 26.44
+        aime2024_accuracy: 10
+        Mathbench_naive_average: 77.08
+        wikibench-wiki-single_choice_cncircular_perf_4: 34
+        cmmlu_naive_average: 75.9
+        mmlu_naive_average: 76.27
+        mmlu_pro_naive_average: 56.14
+        openai_humaneval_humaneval_pass@1: 84.76
+        sanitized_mbpp_score: 74.71
+        humanevalx_naive_average: 48.17
+        ds1000_naive_average: 18.57
+        lcb_code_generation_pass@1: 38.75
+        lcb_code_execution_pass@1: 42.38
+        lcb_test_output_pass@1: 50.45
+        bigcodebench_hard_instruct_pass@1: 16.89
+        bigcodebench_hard_complete_pass@1: 12.16
+        teval_naive_average: 79.46
+        SciCode_sub_accuracy: 100
+        qa_dingo_cn_score: 100
+        mmlu_accuracy: 76.27
+        mmlu-stem_accuracy: 77.75
+        mmlu-social-science_accuracy: 78.65
+        mmlu-humanities_accuracy: 73.12
+        mmlu-other_accuracy: 75.05
+        cmmlu_accuracy: 75.9
+        cmmlu-stem_accuracy: 73.41
+        cmmlu-social-science_accuracy: 75.97
+        cmmlu-humanities_accuracy: 76.42
+        cmmlu-other_accuracy: 78.15
+        cmmlu-china-specific_accuracy: 73.27
+        mmlu_pro_accuracy: 56.14
+        mmlu_pro_biology_accuracy: 72.25
+        mmlu_pro_business_accuracy: 66.16
+        mmlu_pro_chemistry_accuracy: 55.65
+        mmlu_pro_computer_science_accuracy: 60.24
+        mmlu_pro_economics_accuracy: 66.82
+        mmlu_pro_engineering_accuracy: 41.38
+        mmlu_pro_health_accuracy: 54.89
+        mmlu_pro_history_accuracy: 46.46
+        mmlu_pro_law_accuracy: 29.06
+        mmlu_pro_math_accuracy: 73.58
+        mmlu_pro_philosophy_accuracy: 44.89
+        mmlu_pro_physics_accuracy: 60.05
+        mmlu_pro_psychology_accuracy: 61.9
+        mmlu_pro_other_accuracy: 52.6
+        humanevalx-python_pass@1: 51.83
+        humanevalx-cpp_pass@1: 42.68
+        humanevalx-go_pass@1: 0
+        humanevalx-java_pass@1: 73.78
+        humanevalx-js_pass@1: 72.56
+        ds1000_Pandas_accuracy: 14.09
+        ds1000_Numpy_accuracy: 8.64
+        ds1000_Tensorflow_accuracy: 17.78
+        ds1000_Scipy_accuracy: 15.09
+        ds1000_Sklearn_accuracy: 8.7
+        ds1000_Pytorch_accuracy: 4.41
+        ds1000_Matplotlib_accuracy: 61.29
+        mmmlu_lite_accuracy: 54.7
+        openai_mmmlu_lite_AR-XY_accuracy: 42.32
+        openai_mmmlu_lite_BN-BD_accuracy: 42.18
+        openai_mmmlu_lite_DE-DE_accuracy: 60
+        openai_mmmlu_lite_ES-LA_accuracy: 66.18
+        openai_mmmlu_lite_FR-FR_accuracy: 66.88
+        openai_mmmlu_lite_HI-IN_accuracy: 48.63
+        openai_mmmlu_lite_ID-ID_accuracy: 61.26
+        openai_mmmlu_lite_IT-IT_accuracy: 65.26
+        openai_mmmlu_lite_JA-JP_accuracy: 60.7
+        openai_mmmlu_lite_KO-KR_accuracy: 60.63
+        openai_mmmlu_lite_PT-BR_accuracy: 54.46
+        openai_mmmlu_lite_SW-KE_accuracy: 36
+        openai_mmmlu_lite_YO-NG_accuracy: 31.86
+        openai_mmmlu_lite_ZH-CN_accuracy: 69.4
+        college_naive_average: 48.33
+        high_naive_average: 59.33
+        middle_naive_average: 76.67
+        primary_naive_average: 86.67
+        arithmetic_naive_average: 74.33
+        mathbench-a (average)_naive_average: 69.07
+        college_knowledge_naive_average: 83.54
+        high_knowledge_naive_average: 80.82
+        middle_knowledge_naive_average: 83.79
+        primary_knowledge_naive_average: 92.22
+        mathbench-t (average)_naive_average: 85.1
+
+
+internlm3-8b-instruct-turbomind:
+    objective:
+        race-high_accuracy: 89.22
+        ARC-c_accuracy: 92.54
+        BoolQ_accuracy: 86.45
+        triviaqa_wiki_1shot_score: 60.72
+        nq_open_1shot_score: 20.25
+        mmmlu_lite_naive_average: 41.82
+        IFEval_Prompt-level-strict-accuracy: 77.45
+        drop_accuracy: 83.27
+        bbh_naive_average: 55.22
+        GPQA_diamond_accuracy: 37.88
+        hellaswag_accuracy: 91.28
+        TheoremQA_score: 20.12
+        musr_average_naive_average: 36.86
+        korbench_single_naive_average: 41.2
+        ARC_Prize_Public_Evaluation_accuracy: 0.06
+        gsm8k_accuracy: 91.28
+        GaokaoBench_weighted_average: 86.59
+        math_accuracy: 76.96
+        cmo_fib_accuracy: 35.1
+        aime2024_accuracy: 16.67
+        Mathbench_naive_average: 78.96
+        wikibench-wiki-single_choice_cncircular_perf_4: 37.45
+        cmmlu_naive_average: 83.33
+        mmlu_naive_average: 76.21
+        mmlu_pro_naive_average: 57.96
+        openai_humaneval_humaneval_pass@1: 81.71
+        sanitized_mbpp_score: 69.65
+        humanevalx_naive_average: 40.73
+        ds1000_naive_average: 27.23
+        lcb_code_generation_pass@1: 34.75
+        lcb_code_execution_pass@1: 49.9
+        lcb_test_output_pass@1: 48.19
+        bigcodebench_hard_instruct_pass@1: 13.51
+        bigcodebench_hard_complete_pass@1: 15.54
+        teval_naive_average: 82.86
+        SciCode_sub_accuracy: 100
+        qa_dingo_cn_score: 100
+        mmlu_accuracy: 76.21
+        mmlu-stem_accuracy: 77.7
+        mmlu-social-science_accuracy: 80.98
+        mmlu-humanities_accuracy: 70.83
+        mmlu-other_accuracy: 75.01
+        cmmlu_accuracy: 83.33
+        cmmlu-stem_accuracy: 79.66
+        cmmlu-social-science_accuracy: 83.39
+        cmmlu-humanities_accuracy: 84.73
+        cmmlu-other_accuracy: 86.2
+        cmmlu-china-specific_accuracy: 81.77
+        mmlu_pro_accuracy: 57.96
+        mmlu_pro_biology_accuracy: 75.45
+        mmlu_pro_business_accuracy: 64.64
+        mmlu_pro_chemistry_accuracy: 59.81
+        mmlu_pro_computer_science_accuracy: 60.24
+        mmlu_pro_economics_accuracy: 68.6
+        mmlu_pro_engineering_accuracy: 44.79
+        mmlu_pro_health_accuracy: 58.31
+        mmlu_pro_history_accuracy: 49.87
+        mmlu_pro_law_accuracy: 32.43
+        mmlu_pro_math_accuracy: 70.17
+        mmlu_pro_philosophy_accuracy: 46.89
+        mmlu_pro_physics_accuracy: 59.58
+        mmlu_pro_psychology_accuracy: 66.29
+        mmlu_pro_other_accuracy: 54.33
+        humanevalx-python_pass@1: 43.9
+        humanevalx-cpp_pass@1: 20.12
+        humanevalx-go_pass@1: 0
+        humanevalx-java_pass@1: 74.39
+        humanevalx-js_pass@1: 65.24
+        ds1000_Pandas_accuracy: 16.49
+        ds1000_Numpy_accuracy: 34.09
+        ds1000_Tensorflow_accuracy: 26.67
+        ds1000_Scipy_accuracy: 17.92
+        ds1000_Sklearn_accuracy: 20.87
+        ds1000_Pytorch_accuracy: 19.12
+        ds1000_Matplotlib_accuracy: 55.48
+        mmmlu_lite_accuracy: 41.82
+        openai_mmmlu_lite_AR-XY_accuracy: 32.56
+        openai_mmmlu_lite_BN-BD_accuracy: 4.56
+        openai_mmmlu_lite_DE-DE_accuracy: 24.91
+        openai_mmmlu_lite_ES-LA_accuracy: 51.09
+        openai_mmmlu_lite_FR-FR_accuracy: 61.68
+        openai_mmmlu_lite_HI-IN_accuracy: 24.98
+        openai_mmmlu_lite_ID-ID_accuracy: 44.56
+        openai_mmmlu_lite_IT-IT_accuracy: 52.35
+        openai_mmmlu_lite_JA-JP_accuracy: 51.02
+        openai_mmmlu_lite_KO-KR_accuracy: 47.93
+        openai_mmmlu_lite_PT-BR_accuracy: 53.89
+        openai_mmmlu_lite_SW-KE_accuracy: 33.47
+        openai_mmmlu_lite_YO-NG_accuracy: 33.47
+        openai_mmmlu_lite_ZH-CN_accuracy: 69.05
+        college_naive_average: 45.67
+        high_naive_average: 64.67
+        middle_naive_average: 82.33
+        primary_naive_average: 90.33
+        arithmetic_naive_average: 74
+        mathbench-a (average)_naive_average: 71.4
+        college_knowledge_naive_average: 85.28
+        high_knowledge_naive_average: 79.43
+        middle_knowledge_naive_average: 87.9
+        primary_knowledge_naive_average: 93.42
+        mathbench-t (average)_naive_average: 86.51
+
+
+internlm3-8b-instruct-pytorch:
+    objective:
+        race-high_accuracy: 89.02
+        ARC-c_accuracy: 93.56
+        BoolQ_accuracy: 86.67
+        triviaqa_wiki_1shot_score: 60.54
+        nq_open_1shot_score: 20.3
+        mmmlu_lite_naive_average: 42.6
+        IFEval_Prompt-level-strict-accuracy: 79.11
+        drop_accuracy: 83.32
+        bbh_naive_average: 54.76
+        GPQA_diamond_accuracy: 42.42
+        hellaswag_accuracy: 91.31
+        TheoremQA_score: 18
+        musr_average_naive_average: 36.62
+        korbench_single_naive_average: 41.84
+        ARC_Prize_Public_Evaluation_accuracy: 0.06
+        gsm8k_accuracy: 90.67
+        GaokaoBench_weighted_average: 86.27
+        math_accuracy: 76.68
+        cmo_fib_accuracy: 33.65
+        aime2024_accuracy: 10
+        Mathbench_naive_average: 78.92
+        wikibench-wiki-single_choice_cncircular_perf_4: 37.35
+        cmmlu_naive_average: 83.11
+        mmlu_naive_average: 76.23
+        mmlu_pro_naive_average: 58.16
+        openai_humaneval_humaneval_pass@1: 82.32
+        sanitized_mbpp_score: 70.04
+        humanevalx_naive_average: 39.76
+        ds1000_naive_average: 27.84
+        lcb_code_generation_pass@1: 34.5
+        lcb_code_execution_pass@1: 48.02
+        lcb_test_output_pass@1: 47.74
+        bigcodebench_hard_instruct_pass@1: 12.84
+        bigcodebench_hard_complete_pass@1: 15.54
+        teval_naive_average: 82.86
+        SciCode_sub_accuracy: 100
+        qa_dingo_cn_score: 100
+        mmlu_accuracy: 76.23
+        mmlu-stem_accuracy: 78.08
+        mmlu-social-science_accuracy: 80.31
+        mmlu-humanities_accuracy: 71.38
+        mmlu-other_accuracy: 74.63
+        cmmlu_accuracy: 83.11
+        cmmlu-stem_accuracy: 79.42
+        cmmlu-social-science_accuracy: 83.34
+        cmmlu-humanities_accuracy: 83.95
+        cmmlu-other_accuracy: 86.22
+        cmmlu-china-specific_accuracy: 81.5
+        mmlu_pro_accuracy: 58.16
+        mmlu_pro_biology_accuracy: 74.62
+        mmlu_pro_business_accuracy: 65.02
+        mmlu_pro_chemistry_accuracy: 60.69
+        mmlu_pro_computer_science_accuracy: 61.46
+        mmlu_pro_economics_accuracy: 68.25
+        mmlu_pro_engineering_accuracy: 45.3
+        mmlu_pro_health_accuracy: 60.15
+        mmlu_pro_history_accuracy: 50.66
+        mmlu_pro_law_accuracy: 31.7
+        mmlu_pro_math_accuracy: 70.32
+        mmlu_pro_philosophy_accuracy: 47.7
+        mmlu_pro_physics_accuracy: 59.51
+        mmlu_pro_psychology_accuracy: 65.41
+        mmlu_pro_other_accuracy: 53.46
+        humanevalx-python_pass@1: 42.68
+        humanevalx-cpp_pass@1: 19.51
+        humanevalx-go_pass@1: 0
+        humanevalx-java_pass@1: 72.56
+        humanevalx-js_pass@1: 64.02
+        ds1000_Pandas_accuracy: 14.09
+        ds1000_Numpy_accuracy: 35
+        ds1000_Tensorflow_accuracy: 24.44
+        ds1000_Scipy_accuracy: 20.75
+        ds1000_Sklearn_accuracy: 21.74
+        ds1000_Pytorch_accuracy: 22.06
+        ds1000_Matplotlib_accuracy: 56.77
+        mmmlu_lite_accuracy: 42.6
+        openai_mmmlu_lite_AR-XY_accuracy: 32.84
+        openai_mmmlu_lite_BN-BD_accuracy: 10.46
+        openai_mmmlu_lite_DE-DE_accuracy: 24.56
+        openai_mmmlu_lite_ES-LA_accuracy: 50.95
+        openai_mmmlu_lite_FR-FR_accuracy: 61.05
+        openai_mmmlu_lite_HI-IN_accuracy: 30.6
+        openai_mmmlu_lite_ID-ID_accuracy: 45.89
+        openai_mmmlu_lite_IT-IT_accuracy: 51.79
+        openai_mmmlu_lite_JA-JP_accuracy: 51.65
+        openai_mmmlu_lite_KO-KR_accuracy: 48.77
+        openai_mmmlu_lite_PT-BR_accuracy: 52.7
+        openai_mmmlu_lite_SW-KE_accuracy: 32.91
+        openai_mmmlu_lite_YO-NG_accuracy: 32.84
+        openai_mmmlu_lite_ZH-CN_accuracy: 69.33
+        college_naive_average: 47
+        high_naive_average: 66.67
+        middle_naive_average: 81.67
+        primary_naive_average: 89.33
+        arithmetic_naive_average: 73.67
+        mathbench-a (average)_naive_average: 71.67
+        college_knowledge_naive_average: 82.91
+        high_knowledge_naive_average: 79.86
+        middle_knowledge_naive_average: 88.92
+        primary_knowledge_naive_average: 92.96
+        mathbench-t (average)_naive_average: 86.16
diff --git a/.github/scripts/oc_score_baseline_testrange.yaml b/.github/scripts/oc_score_baseline_testrange.yaml
index 45f74131..16a13209 100644
--- a/.github/scripts/oc_score_baseline_testrange.yaml
+++ b/.github/scripts/oc_score_baseline_testrange.yaml
@@ -1,21 +1,24 @@
 chat:
     glm-4-9b-chat-hf:
-        gsm8k_accuracy: 68.75
-        race-high_accuracy: 90.62
+        gsm8k_accuracy: 56.25
+        race-high_accuracy: 84.38
     glm-4-9b-chat-turbomind:
         gsm8k_accuracy: 71.88
         race-high_accuracy: 90.62
     glm-4-9b-chat-vllm:
-        gsm8k_accuracy: 71.88
+        gsm8k_accuracy: 68.75
         race-high_accuracy: 90.62
     deepseek-7b-chat-hf:
         gsm8k_accuracy: 46.88
         race-high_accuracy: 81.25
-    deepseek-moe-16b-chat-hf:
-        gsm8k_accuracy: 50
-        race-high_accuracy: 68.75
+    deepseek-r1-distill-llama-8b-turbomind:
+        gsm8k_accuracy: 31.25
+        race-high_accuracy: 81.25
+    deepseek-r1-distill-qwen-1_5b-turbomind:
+        gsm8k_accuracy: 37.5
+        race-high_accuracy: 53.12
     deepseek-7b-chat-vllm:
-        gsm8k_accuracy: 50
+        gsm8k_accuracy: 43.75
         race-high_accuracy: 78.12
     gemma2-2b-it-hf:
         gsm8k_accuracy: 50
@@ -36,34 +39,40 @@ chat:
         gsm8k_accuracy: 78.12
         race-high_accuracy: 93.75
     gemma-7b-it-vllm:
-        gsm8k_accuracy: 46.88
+        gsm8k_accuracy: 31.25
         race-high_accuracy: 68.75
     internlm2_5-7b-chat-hf:
         gsm8k_accuracy: 84.38
         race-high_accuracy: 90.62
+    internlm3-8b-instruct-hf:
+        gsm8k_accuracy: 65.62
+        race-high_accuracy: 87.5
     internlm2_5-7b-chat-turbomind:
-        gsm8k_accuracy: 87.50
+        gsm8k_accuracy: 84.38
         race-high_accuracy: 90.62
     internlm2-chat-1.8b-turbomind:
         gsm8k_accuracy: 28.12
         race-high_accuracy: 84.38
     internlm2-chat-1.8b-sft-turbomind:
-        gsm8k_accuracy: 21.88
+        gsm8k_accuracy: 31.25
         race-high_accuracy: 84.38
     internlm2-chat-7b-lmdeploy:
-        gsm8k_accuracy: 53.12
+        gsm8k_accuracy: 59.38
         race-high_accuracy: 84.38
     internlm2-chat-7b-sft-turbomind:
-        gsm8k_accuracy: 53.12
+        gsm8k_accuracy: 56.25
         race-high_accuracy: 90.62
+    internlm3-8b-instruct-turbomind:
+        gsm8k_accuracy: 68.75
+        race-high_accuracy: 87.5
     internlm2-chat-7b-vllm:
-        gsm8k_accuracy: 43.75
-        race-high_accuracy: 84.38
+        gsm8k_accuracy: 59.38
+        race-high_accuracy: 87.50
     llama-3_1-8b-instruct-hf:
         gsm8k_accuracy: 84.38
         race-high_accuracy: 90.62
     llama-3_2-3b-instruct-hf:
-        gsm8k_accuracy: 68.75
+        gsm8k_accuracy: 71.88
         race-high_accuracy: 81.25
     llama-3-8b-instruct-hf:
         gsm8k_accuracy: 68.75
@@ -72,14 +81,14 @@ chat:
         gsm8k_accuracy: 18.75
         race-high_accuracy: 46.88
     llama-3_1-8b-instruct-turbomind:
-        gsm8k_accuracy: 78.12
+        gsm8k_accuracy: 81.25
         race-high_accuracy: 90.62
     llama-3_2-3b-instruct-turbomind:
-        gsm8k_accuracy: 65.62
+        gsm8k_accuracy: 75.00
         race-high_accuracy: 81.25
     llama-3-8b-instruct-turbomind:
-        gsm8k_accuracy: 71.88
-        race-high_accuracy: 87.5
+        gsm8k_accuracy: 68.75
+        race-high_accuracy: 84.38
     mistral-7b-instruct-v0.2-hf:
         gsm8k_accuracy: 40.62
         race-high_accuracy: 75
@@ -94,13 +103,10 @@ chat:
         race-high_accuracy: 78.12
     mistral-7b-instruct-v0.1-vllm:
         gsm8k_accuracy: 34.38
-        race-high_accuracy: 68.75
+        race-high_accuracy: 65.62
     mistral-7b-instruct-v0.2-vllm:
-        gsm8k_accuracy: 31.25
-        race-high_accuracy: 75
-    phi-3-mini-4k-instruct-hf:
-        gsm8k_accuracy: 81.25
-        race-high_accuracy: 87.50
+        gsm8k_accuracy: 21.88
+        race-high_accuracy: 78.12
     qwen2.5-0.5b-instruct-hf:
         gsm8k_accuracy: 34.38
         race-high_accuracy: 46.88
@@ -108,10 +114,10 @@ chat:
         gsm8k_accuracy: 53.12
         race-high_accuracy: 90.62
     qwen2.5-0.5b-instruct-turbomind:
-        gsm8k_accuracy: 28.12
-        race-high_accuracy: 50
+        gsm8k_accuracy: 31.25
+        race-high_accuracy: 43.75
     qwen2.5-3b-instruct-turbomind:
-        gsm8k_accuracy: 59.38
+        gsm8k_accuracy: 56.25
         race-high_accuracy: 90.62
     qwen1.5-0.5b-chat-hf:
         gsm8k_accuracy: 0
@@ -123,11 +129,11 @@ chat:
         gsm8k_accuracy: 68.75
         race-high_accuracy: 90.62
     qwen2-1.5b-instruct-turbomind:
-        gsm8k_accuracy: 53.12
+        gsm8k_accuracy: 56.25
         race-high_accuracy: 84.38
     qwen2-7b-instruct-turbomind:
         gsm8k_accuracy: 81.25
-        race-high_accuracy: 90.62
+        race-high_accuracy: 87.50
     qwen1.5-0.5b-chat-vllm:
         gsm8k_accuracy: 3.12
         race-high_accuracy: 53.12
@@ -143,11 +149,11 @@ chat:
     yi-1.5-9b-chat-turbomind:
         gsm8k_accuracy: 71.88
         race-high_accuracy: 93.75
-    deepseek-v2-lite-chat-hf:
-        gsm8k_accuracy: 46.88
+    deepseek-v2_lite-chat-turbomind:
+        gsm8k_accuracy: 37.5
         race-high_accuracy: 71.88
     gemma2-27b-it-hf:
-        gsm8k_accuracy: 75
+        gsm8k_accuracy: 71.88
         race-high_accuracy: 93.75
     internlm2_5-20b-chat-hf:
         gsm8k_accuracy: 84.38
@@ -161,6 +167,9 @@ chat:
     mistral-small-instruct-2409-turbomind:
         gsm8k_accuracy: 81.25
         race-high_accuracy: 87.50
+    phi-4:
+        gsm8k_accuracy: 81.25
+        race-high_accuracy: 87.50
     qwen2.5-14b-instruct-hf:
         gsm8k_accuracy: 71.88
         race-high_accuracy: 96.88
@@ -168,40 +177,41 @@ chat:
         gsm8k_accuracy: 68.75
         race-high_accuracy: 93.75
     yi-1.5-34b-chat-turbomind:
-        gsm8k_accuracy: 78.12
+        gsm8k_accuracy: 75.00
         race-high_accuracy: 93.75
-    deepseek-67b-chat-hf:
-        gsm8k_accuracy: 71.88
+    deepseek-67b-chat-turbomind:
+        gsm8k_accuracy: 75.00
         race-high_accuracy: 78.12
+    deepseek-r1-distill-qwen-32b-turbomind:
+        gsm8k_accuracy: 25
+        race-high_accuracy: 90.62
     llama-3_3-70b-instruct-turbomind:
         gsm8k_accuracy: 93.75
         race-high_accuracy: 87.5
-    mixtral-8x7b-instruct-v0.1-hf:
-        gsm8k_accuracy: 59.38
-        race-high_accuracy: 81.25
     mixtral-large-instruct-2411-turbomind:
-        gsm8k_accuracy: 90.62
+        gsm8k_accuracy: 87.50
         race-high_accuracy: 93.75
     nvidia-3_1-Nemotron-70b-instruct-HF-turbomind:
-        gsm8k_accuracy: 87.5
-        race-high_accuracy: 46.88
+        gsm8k_accuracy: 93.75
+        race-high_accuracy: 50.00
     qwen2.5-72b-instruct-turbomind:
-        gsm8k_accuracy: 75
-        race-high_accuracy: 93.75
+        gsm8k_accuracy: 81.25
+        race-high_accuracy: 90.62
+    deepseek-r1-distill-llama-70b-turbomind:
+        gsm8k_accuracy: 40.62
+        race-high_accuracy: 90.62
     deepseek-v2_5-1210-turbomind:
         gsm8k_accuracy: 90.62
         race-high_accuracy: 84.38
-    mixtral-8x22b-instruct-v0.1-hf:
-        gsm8k_accuracy: 81.25
-        race-high_accuracy: 81.25
+    mixtral-8x22b-instruct-v0.1-turbomind:
+        gsm8k_accuracy: 75
+        race-high_accuracy: 78.12
+    mixtral-8x22b-instruct-v0.1-vllm:
+        gsm8k_accuracy: 78.12
+        race-high_accuracy: 78.12
 base:
-    glm-4-9b-hf:
-        gsm8k_accuracy: 68.75
-        GPQA_diamond_accuracy: 31.25
-        race-high_accuracy: 93.75
-        winogrande_accuracy: 84.38
     glm-4-9b-turbomind:
-        gsm8k_accuracy: 62.5
+        gsm8k_accuracy: 56.25
         GPQA_diamond_accuracy: 28.12
         race-high_accuracy: 93.75
         winogrande_accuracy: 84.38
@@ -210,15 +220,10 @@ base:
         GPQA_diamond_accuracy: 0
         race-high_accuracy: 46.88
         winogrande_accuracy: 71.88
-    deepseek-moe-16b-base-hf:
-        gsm8k_accuracy: 21.88
-        GPQA_diamond_accuracy: 0
-        race-high_accuracy: 21.88
-        winogrande_accuracy: 65.62
     deepseek-7b-base-turbomind:
-        gsm8k_accuracy: 21.88
+        gsm8k_accuracy: 18.75
         GPQA_diamond_accuracy: 0
-        race-high_accuracy: 46.88
+        race-high_accuracy: 43.75
         winogrande_accuracy: 84.38
     deepseek-moe-16b-base-vllm:
         gsm8k_accuracy: 21.88
@@ -245,16 +250,21 @@ base:
         GPQA_diamond_accuracy: 3.12
         race-high_accuracy: 65.62
         winogrande_accuracy: 71.88
+    gemma-2-9b-turbomind:
+        gsm8k_accuracy: 68.75
+        GPQA_diamond_accuracy: 0
+        race-high_accuracy: 78.12
+        winogrande_accuracy: 50
     gemma-2b-vllm:
         gsm8k_accuracy: 15.62
         GPQA_diamond_accuracy: 3.12
-        race-high_accuracy:
-        winogrande_accuracy:
+        race-high_accuracy: 28.12
+        winogrande_accuracy: 68.75
     gemma-7b-vllm:
-        gsm8k_accuracy: 53.12
-        GPQA_diamond_accuracy: 9.38
-        race-high_accuracy:
-        winogrande_accuracy:
+        gsm8k_accuracy: 43.75
+        GPQA_diamond_accuracy: 6.25
+        race-high_accuracy: 81.25
+        winogrande_accuracy: 81.25
     internlm2_5-7b-hf:
         gsm8k_accuracy: 37.5
         GPQA_diamond_accuracy: 25
@@ -265,30 +275,25 @@ base:
         GPQA_diamond_accuracy: 18.75
         race-high_accuracy: 62.5
         winogrande_accuracy: 78.12
-    internlm2-base-7b-hf:
-        gsm8k_accuracy: 3.12
-        GPQA_diamond_accuracy: 21.88
-        race-high_accuracy: 75
-        winogrande_accuracy: 65.62
     internlm2-1.8b-turbomind:
-        gsm8k_accuracy: 12.5
-        GPQA_diamond_accuracy: 9.38
+        gsm8k_accuracy: 6.25
+        GPQA_diamond_accuracy: 12.5
         race-high_accuracy: 71.88
-        winogrande_accuracy: 78.12
+        winogrande_accuracy: 75
     internlm2_5-7b-turbomind:
-        gsm8k_accuracy: 62.50
+        gsm8k_accuracy: 59.38
         GPQA_diamond_accuracy: 34.38
         race-high_accuracy: 93.75
-        winogrande_accuracy: 87.50
+        winogrande_accuracy: 84.38
     internlm2-7b-turbomind:
-        gsm8k_accuracy: 53.12
-        GPQA_diamond_accuracy: 21.88
+        gsm8k_accuracy: 50
+        GPQA_diamond_accuracy: 18.75
         race-high_accuracy: 71.88
         winogrande_accuracy: 84.38
     internlm2-base-7b-turbomind:
         gsm8k_accuracy: 37.50
-        GPQA_diamond_accuracy: 28.12
-        race-high_accuracy: 81.25
+        GPQA_diamond_accuracy: 21.88
+        race-high_accuracy: 84.38
         winogrande_accuracy: 75
     llama-2-7b-hf:
         gsm8k_accuracy: 21.88
@@ -311,7 +316,7 @@ base:
         race-high_accuracy: 78.12
         winogrande_accuracy: 78.12
     llama-3-8b-turbomind:
-        gsm8k_accuracy: 50
+        gsm8k_accuracy: 46.88
         GPQA_diamond_accuracy: 12.50
         race-high_accuracy: 65.62
         winogrande_accuracy: 78.12
@@ -327,14 +332,14 @@ base:
         winogrande_accuracy: 71.88
     qwen2.5-1.5b-turbomind:
         gsm8k_accuracy: 62.50
-        GPQA_diamond_accuracy: 12.50
-        race-high_accuracy: 78.12
-        winogrande_accuracy: 68.75
-    qwen2.5-7b-turbomind:
-        gsm8k_accuracy: 75.00
-        GPQA_diamond_accuracy: 25
-        race-high_accuracy: 87.5
+        GPQA_diamond_accuracy: 15.62
+        race-high_accuracy: 75
         winogrande_accuracy: 71.88
+    qwen2.5-7b-turbomind:
+        gsm8k_accuracy: 71.88
+        GPQA_diamond_accuracy: 18.75
+        race-high_accuracy: 87.5
+        winogrande_accuracy: 75.00
     qwen1.5-moe-a2.7b-hf:
         gsm8k_accuracy: 62.5
         GPQA_diamond_accuracy: 18.75
@@ -356,17 +361,17 @@ base:
         race-high_accuracy: 87.5
         winogrande_accuracy: 68.75
     qwen2-1.5b-turbomind:
-        gsm8k_accuracy: 56.25
-        GPQA_diamond_accuracy: 9.38
+        gsm8k_accuracy: 59.38
+        GPQA_diamond_accuracy: 12.50
         race-high_accuracy: 81.25
         winogrande_accuracy: 75
     qwen2-7b-turbomind:
-        gsm8k_accuracy: 75.00
+        gsm8k_accuracy: 65.62
         GPQA_diamond_accuracy: 12.5
         race-high_accuracy: 87.5
         winogrande_accuracy: 71.88
     qwen1.5-0.5b-vllm:
-        gsm8k_accuracy: 9.38
+        gsm8k_accuracy: 6.25
         GPQA_diamond_accuracy: 0
         race-high_accuracy: 56.25
         winogrande_accuracy: 62.5
@@ -382,27 +387,12 @@ base:
         winogrande_accuracy: 59.38
     yi-1.5-9b-turbomind:
         gsm8k_accuracy: 78.12
-        GPQA_diamond_accuracy: 40.62
+        GPQA_diamond_accuracy: 43.75
         race-high_accuracy: 87.5
         winogrande_accuracy: 71.88
-    deepseek-v2-lite-hf:
-        gsm8k_accuracy: 31.25
-        GPQA_diamond_accuracy: 28.12
-        race-high_accuracy: 59.38
-        winogrande_accuracy: 71.88
-    internlm2-20b-hf:
-        gsm8k_accuracy: 56.25
-        GPQA_diamond_accuracy: 15.62
-        race-high_accuracy: 68.75
-        winogrande_accuracy: 75
-    internlm2-base-20b-hf:
-        gsm8k_accuracy: 12.5
-        GPQA_diamond_accuracy: 9.38
-        race-high_accuracy: 84.38
-        winogrande_accuracy: 65.62
     internlm2-20b-turbomind:
-        gsm8k_accuracy: 71.88
-        GPQA_diamond_accuracy: 15.62
+        gsm8k_accuracy: 75
+        GPQA_diamond_accuracy: 18.75
         race-high_accuracy: 68.75
         winogrande_accuracy: 81.25
     qwen2.5-14b-hf:
@@ -416,37 +406,27 @@ base:
         race-high_accuracy: 93.75
         winogrande_accuracy: 78.12
     qwen2.5-32b-turbomind:
-        gsm8k_accuracy: 84.38
-        GPQA_diamond_accuracy: 28.12
+        gsm8k_accuracy: 87.5
+        GPQA_diamond_accuracy: 18.75
         race-high_accuracy: 93.75
         winogrande_accuracy: 81.25
-    deepseek-67b-base-hf:
-        gsm8k_accuracy: 59.38
-        GPQA_diamond_accuracy: 31.25
-        race-high_accuracy: 81.25
-        winogrande_accuracy: 90.62
     deepseek-67b-base-turbomind:
-        gsm8k_accuracy: 56.25
+        gsm8k_accuracy: 53.12
         GPQA_diamond_accuracy: 28.12
         race-high_accuracy: 81.25
         winogrande_accuracy: 84.38
     llama-3-70b-turbomind:
-        gsm8k_accuracy: 59.38
-        GPQA_diamond_accuracy: 9.38
+        gsm8k_accuracy: 56.25
+        GPQA_diamond_accuracy: 12.50
         race-high_accuracy: 93.75
         winogrande_accuracy: 84.38
     qwen2.5-72b-turbomind:
         gsm8k_accuracy: 84.38
-        GPQA_diamond_accuracy: 34.38
+        GPQA_diamond_accuracy: 31.25
         race-high_accuracy: 93.75
         winogrande_accuracy: 87.5
     deepseek-v2-turbomind:
-        gsm8k_accuracy: 65.62
-        GPQA_diamond_accuracy: 15.62
-        race-high_accuracy: 93.75
-        winogrande_accuracy: 84.38
-    llama-3-70b-hf:
-        gsm8k_accuracy: 62.5
+        gsm8k_accuracy: 59.38
         GPQA_diamond_accuracy: 3.12
         race-high_accuracy: 93.75
-        winogrande_accuracy: 84.38
+        winogrande_accuracy: 81.25
diff --git a/.github/workflows/daily-run-test.yml b/.github/workflows/daily-run-test.yml
index a5a930fa..6a1c2ebc 100644
--- a/.github/workflows/daily-run-test.yml
+++ b/.github/workflows/daily-run-test.yml
@@ -61,6 +61,7 @@ env:
   HUGGINGFACE_HUB_CACHE: /fs-computility/llm/shared/llmeval/models/opencompass_hf_hub
   HF_HUB_CACHE: /fs-computility/llm/shared/llmeval/models/opencompass_hf_hub
   CONDA_ENV: regression_test
+  export VLLM_WORKER_MULTIPROC_METHOD: spawn
 
 jobs:
   build-pypi:
@@ -92,7 +93,6 @@ jobs:
       matrix:
         pyver: [py310]
     runs-on: ubuntu-latest
-    environment: 'prod'
     env:
       PYTHON_VERSION: ${{ matrix.pyver }}
       PLAT_NAME: manylinux2014_x86_64
@@ -126,7 +126,6 @@ jobs:
     if: ${{!cancelled()}}
     needs: ['build-pypi', 'build-pypi-lmdeploy']
     runs-on: volc_cu12
-    environment: 'prod'
     timeout-minutes: 120 #2hours
     steps:
       - name: Clone repository
@@ -190,7 +189,6 @@ jobs:
       matrix:
         regression_func: ${{fromJSON(github.event.inputs.regression_func_volc || '["chat_models","base_models","chat_obj_fullbench","base_fullbench"]')}}
     runs-on: volc_cu12_daily
-    environment: 'prod'
     timeout-minutes: 180 #3hours
     steps:
       - name: Clone repository
@@ -231,7 +229,6 @@ jobs:
       matrix:
         regression_func: ${{fromJSON(github.event.inputs.regression_func_local || '["cmd","api","chat_sub_fullbench"]')}}
     runs-on: volc_cu12_local
-    environment: 'prod'
     timeout-minutes: 480 #6hours
     steps:
       - name: Clone repository
@@ -258,27 +255,33 @@ jobs:
           conda info --envs
           export from_tf=TRUE
           python tools/list_configs.py internlm2_5 mmlu
-          opencompass --models hf_internlm2_5_7b hf_internlm2_1_8b --datasets race_ppl demo_gsm8k_chat_gen --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd1 --reuse --max-num-workers 2 --dump-eval-details
+          opencompass --models hf_internlm2_5_7b --datasets race_ppl demo_gsm8k_chat_gen --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd1 --reuse --max-num-workers 2 --dump-eval-details
           rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd1/*/summary regression_result_daily
           python -m pytest -m case1 -s -v --color=yes .github/scripts/oc_score_assert.py
-          opencompass --models hf_internlm2_5_7b_chat hf_internlm2_chat_1_8b --datasets race_gen demo_gsm8k_chat_gen -a lmdeploy --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd2 --reuse --max-num-workers 2 --dump-eval-details
+          opencompass --models hf_internlm2_5_7b_chat hf_internlm3_8b_instruct --datasets race_gen demo_gsm8k_chat_gen -a lmdeploy --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd2 --reuse --max-num-workers 2 --dump-eval-details
           rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd2/*/summary regression_result_daily
           python -m pytest -m case2 -s -v --color=yes .github/scripts/oc_score_assert.py
           opencompass --datasets race_ppl demo_gsm8k_chat_gen --hf-type base --hf-path internlm/internlm2_5-7b --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd3 --reuse --max-num-workers 2 --dump-eval-details
           rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd3/*/summary regression_result_daily
           python -m pytest -m case3 -s -v --color=yes .github/scripts/oc_score_assert.py
-          opencompass --datasets race_gen demo_gsm8k_chat_gen --hf-type chat --hf-path internlm/internlm2_5-7b-chat --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd4 --reuse --max-num-workers 2 --dump-eval-details
+          opencompass --datasets race_gen demo_gsm8k_chat_gen --hf-type chat --hf-path internlm/internlm3-8b-instruct -a lmdeploy --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd4 --reuse --max-num-workers 2 --dump-eval-details
           rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd4/*/summary regression_result_daily
           python -m pytest -m case4 -s -v --color=yes .github/scripts/oc_score_assert.py
+          opencompass --datasets race_gen demo_gsm8k_chat_gen --hf-type chat --hf-path internlm/internlm3-8b-instruct -a vllm --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd5 --reuse --max-num-workers 2 --dump-eval-details
+          rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/cmd5/*/summary regression_result_daily
+          python -m pytest -m case5 -s -v --color=yes .github/scripts/oc_score_assert.py
       - name:  Run model test - api
         if: matrix.regression_func == 'api'
         run: |
           . ${{env.CONDA_PATH}}/bin/activate
           conda activate ${{env.CONDA_ENV}}
           conda info --envs
-          lmdeploy serve api_server internlm/internlm2_5-7b-chat --max-batch-size 256 --model-name internlm2 > ${{env.REPORT_ROOT}}/${{ github.run_id }}/restful.log  2>&1  &
+          lmdeploy serve api_server internlm/internlm3-8b-instruct --max-batch-size 256 --model-name internlm3 > ${{env.REPORT_ROOT}}/${{ github.run_id }}/restful.log  2>&1  &
           echo "restful_pid=$!" >> "$GITHUB_ENV"
           sleep 180s
+          env | grep PROXY
+          env | grep proxy
+          unset HTTP_PROXY;unset HTTPS_PROXY;unset http_proxy;unset https_proxy;
           opencompass .github/scripts/eval_regression_api.py --work-dir ${{env.REPORT_ROOT}}/${{ github.run_id }}/api --reuse --max-num-workers 2 --dump-eval-details
           rm regression_result_daily -f && ln -s ${{env.REPORT_ROOT}}/${{ github.run_id }}/api/*/summary regression_result_daily
           python -m pytest -m api -s -v --color=yes .github/scripts/oc_score_assert.py
@@ -307,7 +310,6 @@ jobs:
       matrix:
         function_type: ${{fromJSON(github.event.inputs.fullbench_eval || '["base_objective","chat_objective","chat_subjective","base_long_context","chat_long_context"]')}}
     runs-on: volc_cu12
-    environment: 'prod'
     timeout-minutes: 480 #6hours
     steps:
       - name: Clone repository
@@ -341,7 +343,6 @@ jobs:
     needs: [daily_run_test_volc, daily_run_test_local, fullbench_run_test]
     timeout-minutes: 5
     runs-on: self-hosted
-    environment: 'prod'
     steps:
       - name: notify
         run: |
diff --git a/opencompass/configs/models/mistral/lmdeploy_mixtral_8x22b_instruct_v0_1.py b/opencompass/configs/models/mistral/lmdeploy_mixtral_8x22b_instruct_v0_1.py
new file mode 100644
index 00000000..1ffef256
--- /dev/null
+++ b/opencompass/configs/models/mistral/lmdeploy_mixtral_8x22b_instruct_v0_1.py
@@ -0,0 +1,22 @@
+from opencompass.models import TurboMindModelwithChatTemplate
+
+models = [
+    dict(
+        type=TurboMindModelwithChatTemplate,
+        abbr='mixtral-8x22b-instruct-v0.1-turbomind',
+        path='mistralai/Mixtral-8x22B-Instruct-v0.1',
+        engine_config=dict(
+            session_len=32768,
+            max_batch_size=16,
+            tp=8,
+            cache_max_entry_count=0.7,
+        ),
+        gen_config=dict(
+            top_k=1, temperature=1e-6, top_p=0.9, max_new_tokens=4096
+        ),
+        max_seq_len=32768,
+        max_out_len=4096,
+        batch_size=8,
+        run_cfg=dict(num_gpus=8),
+    )
+]
diff --git a/opencompass/summarizers/subjective/common_summarizer.py b/opencompass/summarizers/subjective/common_summarizer.py
index ccb8d139..de917f44 100644
--- a/opencompass/summarizers/subjective/common_summarizer.py
+++ b/opencompass/summarizers/subjective/common_summarizer.py
@@ -147,6 +147,5 @@ class CommonSummarizer(CompassArenaSummarizer):
             f.write(','.join(new_header) + '\n')
             for line in new_table:
                 f.write(','.join(map(str, line)) + '\n')
-            print(t)
             print(output_file)
         return {'qa_bench_' + show_dataset_abbr:json_result}