[ci] refactor testcoverage config by zhulinJulia24 · Pull Request #4630 · InternLM/lmdeploy

zhulinJulia24 · 2026-05-28T05:40:33Z

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily receiving feedbacks. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Please describe the motivation of this PR and the goal you want to achieve through this PR.

Modification

Please briefly describe what modification is made in this PR.

BC-breaking (Optional)

Does the modification introduce changes that break the backward-compatibility of the downstream repositories?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

Pre-commit or other linting tools are used to fix the potential lint issues.
The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
The documentation has been modified accordingly, like docstring or example tutorials.

Copilot

Pull request overview

This PR refactors the autotest “test coverage” configuration and CI wiring by moving environment-specific path settings into a single autotest/env_paths.yml, switching evaluation to a unified preset resolver, and introducing/normalizing many per-model config matrices under autotest/configs/**. It also updates several CI workflows to use a DEPS_PROFILE matrix for pinned dependency rows and adjusts a few benchmark/test utilities accordingly.

Changes:

Centralize environment path configuration in autotest/env_paths.yml and update autotest fixtures to load config via get_config().
Route API evaluation preset selection through get_eval_preset_config(...) (enabling per-model gen_config overrides while keeping scalar preset limits).
Add/normalize a large set of per-model autotest config YAMLs and update CI workflows (checkout version bumps + new DEPS_PROFILE strategy).

Reviewed changes

Copilot reviewed 144 out of 144 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
autotest/utils/constant.py	Adds clarifying comments and switches “legacy” selection logic to `DEPS_PROFILE`-based detection.
autotest/utils/benchmark_utils.py	Adds `--trust-remote-code` for restful profiling and refactors prefix-cache throughput command construction.
autotest/tools/common_case_config.py	Updates fallback test model choices (notably for Turbomind MLLM fallback).
autotest/interface/restful/test_restful_completions_v1.py	Adds extensive debug `print(...)` output in RESTful completions tests.
autotest/interface/restful/test_restful_chat_completions_v1.py	Relaxes empty-content assertions to accept falsy content.
autotest/interface/pipeline/test_pipeline_longtext_func.py	Refactors longtext tests: adds spawn exitcode check, adds response assertions, introduces stream-merge helpers, updates model list.
autotest/evaluate/test_mllm_api_evaluate.py	Uses `get_eval_preset_config(..., mllm=True)` instead of direct constants lookup.
autotest/evaluate/test_api_evaluate.py	Uses `get_eval_preset_config(...)` instead of direct constants lookup.
autotest/evaluate/eval_config_chat.py	Reduces evaluation batch size (1024 → 256).
autotest/env_paths.yml	New: central per-environment paths for autotest runs.
autotest/conftest.py	Switches session config loading to `get_config()` (env_paths-driven).
autotest/configs/zai-org/GLM-5-FP8.yml	New per-model matrix entry for `zai-org/GLM-5-FP8`.
autotest/configs/zai-org/GLM-4.7-Flash.yml	New per-model matrix entry for `zai-org/GLM-4.7-Flash`.
autotest/configs/unsloth/gpt-oss-20b-BF16.yml	New per-model matrix entry for `unsloth/gpt-oss-20b-BF16`.
autotest/configs/THUDM/glm-4v-9b.yml	New per-model matrix entry for `THUDM/glm-4v-9b`.
autotest/configs/THUDM/glm-4-9b-chat.yml	New per-model matrix entry for `THUDM/glm-4-9b-chat`.
autotest/configs/THUDM/cogvlm2-llama3-chinese-chat-19B.yml	New per-model matrix entry for `THUDM/cogvlm2-llama3-chinese-chat-19B`.
autotest/configs/THUDM/cogvlm-chat-hf.yml	New per-model matrix entry for `THUDM/cogvlm-chat-hf`.
autotest/configs/THUDM/chatglm2-6b.yml	New per-model matrix entry for `THUDM/chatglm2-6b`.
autotest/configs/test/test_vl_tp8.yml	New test matrix config for `test_vl_tp8`.
autotest/configs/test/test_vl_tp2.yml	New test matrix config for `test_vl_tp2`.
autotest/configs/test/test_vl_tp2_int4.yml	New test matrix config for `test_vl_tp2_int4`.
autotest/configs/test/test_vl_tp2_gpqa.yml	New test matrix config for `test_vl_tp2_gpqa`.
autotest/configs/test/test_vl_tp1.yml	New test matrix config for `test_vl_tp1`.
autotest/configs/test/test_vl_tp1_pytorch.yml	New test matrix config for `test_vl_tp1_pytorch`.
autotest/configs/test/test_vl_dpep8.yml	New test matrix config for `test_vl_dpep8`.
autotest/configs/test/test_vl_dpep16.yml	New test matrix config for `test_vl_dpep16`.
autotest/configs/test/test_vl_cp2tp8.yml	New test matrix config for `test_vl_cp2tp8`.
autotest/configs/test/test_tp8.yml	New test matrix config for `test_tp8`.
autotest/configs/test/test_tp2.yml	New test matrix config for `test_tp2`.
autotest/configs/test/test_tp2_int4.yml	New test matrix config for `test_tp2_int4`.
autotest/configs/test/test_tp2_gpqa.yml	New test matrix config for `test_tp2_gpqa`.
autotest/configs/test/test_tp1.yml	New test matrix config for `test_tp1`.
autotest/configs/test/test_tp1_pytorch.yml	New test matrix config for `test_tp1_pytorch`.
autotest/configs/test/test_dpep8.yml	New test matrix config for `test_dpep8`.
autotest/configs/test/test_dpep16.yml	New test matrix config for `test_dpep16`.
autotest/configs/test/test_cp2tp8.yml	New test matrix config for `test_cp2tp8`.
autotest/configs/test/test_allkind.yml	New test matrix config for `test_allkind`.
autotest/configs/Qwen/Qwen3.5-397B-A17B.yml	New per-model matrix entry for `Qwen/Qwen3.5-397B-A17B`.
autotest/configs/Qwen/Qwen3.5-397B-A17B-FP8.yml	New per-model matrix entry for `Qwen/Qwen3.5-397B-A17B-FP8`.
autotest/configs/Qwen/Qwen3.5-35B-A3B.yml	New per-model matrix entry for `Qwen/Qwen3.5-35B-A3B`.
autotest/configs/Qwen/Qwen3.5-35B-A3B-FP8.yml	New per-model matrix entry for `Qwen/Qwen3.5-35B-A3B-FP8`.
autotest/configs/Qwen/Qwen3.5-35B-A3B-Base.yml	New per-model matrix entry for `Qwen/Qwen3.5-35B-A3B-Base`.
autotest/configs/Qwen/Qwen3.5-2B-Base.yml	New per-model matrix entry for `Qwen/Qwen3.5-2B-Base`.
autotest/configs/Qwen/Qwen3.5-27B.yml	New per-model matrix entry for `Qwen/Qwen3.5-27B`.
autotest/configs/Qwen/Qwen3.5-122B-A10B.yml	New per-model matrix entry for `Qwen/Qwen3.5-122B-A10B`.
autotest/configs/Qwen/Qwen3.5-0.8B.yml	New per-model matrix entry for `Qwen/Qwen3.5-0.8B`.
autotest/configs/Qwen/Qwen3-VL-8B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen3-VL-8B-Instruct`.
autotest/configs/Qwen/Qwen3-VL-32B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen3-VL-32B-Instruct`.
autotest/configs/Qwen/Qwen3-VL-30B-A3B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen3-VL-30B-A3B-Instruct`.
autotest/configs/Qwen/Qwen3-8B.yml	New per-model matrix entry for `Qwen/Qwen3-8B`.
autotest/configs/Qwen/Qwen3-8B-Base.yml	New per-model matrix entry for `Qwen/Qwen3-8B-Base`.
autotest/configs/Qwen/Qwen3-4B.yml	New per-model matrix entry for `Qwen/Qwen3-4B`.
autotest/configs/Qwen/Qwen3-32B.yml	New per-model matrix entry for `Qwen/Qwen3-32B`.
autotest/configs/Qwen/Qwen3-32B-FP8.yml	New per-model matrix entry for `Qwen/Qwen3-32B-FP8`.
autotest/configs/Qwen/Qwen3-30B-A3B.yml	New per-model matrix entry for `Qwen/Qwen3-30B-A3B`.
autotest/configs/Qwen/Qwen3-30B-A3B-GPTQ-Int4.yml	New per-model matrix entry for `Qwen/Qwen3-30B-A3B-GPTQ-Int4`.
autotest/configs/Qwen/Qwen3-30B-A3B-FP8.yml	New per-model matrix entry for `Qwen/Qwen3-30B-A3B-FP8`.
autotest/configs/Qwen/Qwen3-30B-A3B-Base.yml	New per-model matrix entry for `Qwen/Qwen3-30B-A3B-Base`.
autotest/configs/Qwen/Qwen3-235B-A22B.yml	New per-model matrix entry for `Qwen/Qwen3-235B-A22B`.
autotest/configs/Qwen/Qwen3-235B-A22B-Thinking-2507.yml	New per-model matrix entry for `Qwen/Qwen3-235B-A22B-Thinking-2507`.
autotest/configs/Qwen/Qwen3-235B-A22B-Thinking-2507-FP8.yml	New per-model matrix entry for `Qwen/Qwen3-235B-A22B-Thinking-2507-FP8`.
autotest/configs/Qwen/Qwen3-1.7B.yml	New per-model matrix entry for `Qwen/Qwen3-1.7B`.
autotest/configs/Qwen/Qwen3-0.6B.yml	New per-model matrix entry for `Qwen/Qwen3-0.6B`.
autotest/configs/Qwen/Qwen2.5-VL-7B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen2.5-VL-7B-Instruct`.
autotest/configs/Qwen/Qwen2.5-VL-3B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen2.5-VL-3B-Instruct`.
autotest/configs/Qwen/Qwen2.5-VL-32B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen2.5-VL-32B-Instruct`.
autotest/configs/Qwen/Qwen2.5-7B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen2.5-7B-Instruct`.
autotest/configs/Qwen/Qwen2.5-72B-Instruct.yml	New per-model matrix entry for `Qwen/Qwen2.5-72B-Instruct`.
autotest/configs/Qwen/Qwen2-57B-A14B-Instruct-GPTQ-Int4.yml	New per-model matrix entry for `Qwen/Qwen2-57B-A14B-Instruct-GPTQ-Int4`.
autotest/configs/Qwen/Qwen1.5-MoE-A2.7B-Chat.yml	New per-model matrix entry for `Qwen/Qwen1.5-MoE-A2.7B-Chat`.
autotest/configs/OpenGVLab/Mini-InternVL-Chat-2B-V1-5.yml	New per-model matrix entry for `OpenGVLab/Mini-InternVL-Chat-2B-V1-5`.
autotest/configs/OpenGVLab/InternVL3-8B.yml	New per-model matrix entry for `OpenGVLab/InternVL3-8B`.
autotest/configs/OpenGVLab/InternVL3-38B.yml	New per-model matrix entry for `OpenGVLab/InternVL3-38B`.
autotest/configs/OpenGVLab/InternVL3-2B-Instruct.yml	New per-model matrix entry for `OpenGVLab/InternVL3-2B-Instruct`.
autotest/configs/OpenGVLab/InternVL3-1B-Instruct.yml	New per-model matrix entry for `OpenGVLab/InternVL3-1B-Instruct`.
autotest/configs/OpenGVLab/InternVL3_5-8B.yml	New per-model matrix entry for `OpenGVLab/InternVL3_5-8B`.
autotest/configs/OpenGVLab/InternVL3_5-38B.yml	New per-model matrix entry for `OpenGVLab/InternVL3_5-38B`.
autotest/configs/OpenGVLab/InternVL3_5-30B-A3B.yml	New per-model matrix entry for `OpenGVLab/InternVL3_5-30B-A3B`.
autotest/configs/OpenGVLab/InternVL3_5-2B.yml	New per-model matrix entry for `OpenGVLab/InternVL3_5-2B`.
autotest/configs/OpenGVLab/InternVL2-Llama3-76B-AWQ.yml	New per-model matrix entry for `OpenGVLab/InternVL2-Llama3-76B-AWQ`.
autotest/configs/OpenGVLab/InternVL2_5-8B.yml	New per-model matrix entry for `OpenGVLab/InternVL2_5-8B`.
autotest/configs/OpenGVLab/InternVL2_5-1B.yml	New per-model matrix entry for `OpenGVLab/InternVL2_5-1B`.
autotest/configs/openai/gpt-oss-20b.yml	New per-model matrix entry for `openai/gpt-oss-20b`.
autotest/configs/openai/gpt-oss-120b.yml	New per-model matrix entry for `openai/gpt-oss-120b`.
autotest/configs/moonshotai/Kimi-K2-Instruct-0905.yml	New per-model matrix entry for `moonshotai/Kimi-K2-Instruct-0905`.
autotest/configs/mistralai/Mixtral-8x7B-Instruct-v0.1.yml	New per-model matrix entry for `mistralai/Mixtral-8x7B-Instruct-v0.1`.
autotest/configs/microsoft/Phi-4-mini-instruct.yml	New per-model matrix entry for `microsoft/Phi-4-mini-instruct`.
autotest/configs/meta-llama/Meta-Llama-3.1-8B-Instruct.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3.1-8B-Instruct`.
autotest/configs/meta-llama/Meta-Llama-3.1-70B-Instruct.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3.1-70B-Instruct`.
autotest/configs/meta-llama/Meta-Llama-3-8B-Instruct.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3-8B-Instruct`.
autotest/configs/meta-llama/Meta-Llama-3-1-8B-Instruct.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3-1-8B-Instruct`.
autotest/configs/meta-llama/Meta-Llama-3-1-8B-Instruct-AWQ.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3-1-8B-Instruct-AWQ`.
autotest/configs/meta-llama/Meta-Llama-3-1-70B-Instruct.yml	New per-model matrix entry for `meta-llama/Meta-Llama-3-1-70B-Instruct`.
autotest/configs/meta-llama/Llama-4-Scout-17B-16E-Instruct.yml	New per-model matrix entry for `meta-llama/Llama-4-Scout-17B-16E-Instruct`.
autotest/configs/meta-llama/Llama-3.2-3B-Instruct.yml	New per-model matrix entry for `meta-llama/Llama-3.2-3B-Instruct`.
autotest/configs/meta-llama/Llama-3.2-1B-Instruct.yml	New per-model matrix entry for `meta-llama/Llama-3.2-1B-Instruct`.
autotest/configs/meta-llama/Llama-2-7b-chat-hf.yml	New per-model matrix entry for `meta-llama/Llama-2-7b-chat-hf`.
autotest/configs/liuhaotian/llava-v1.6-vicuna-7b.yml	New per-model matrix entry for `liuhaotian/llava-v1.6-vicuna-7b`.
autotest/configs/JetLM/SDAR-30B-A3B-Sci.yml	New per-model matrix entry for `JetLM/SDAR-30B-A3B-Sci`.
autotest/configs/internlm/internlm3-8b-instruct.yml	New per-model matrix entry for `internlm/internlm3-8b-instruct`.
autotest/configs/internlm/internlm3-8b-instruct-awq.yml	New per-model matrix entry for `internlm/internlm3-8b-instruct-awq`.
autotest/configs/internlm/Intern-S1.yml	New per-model matrix entry for `internlm/Intern-S1`.
autotest/configs/internlm/Intern-S1-Pro-FP8.yml	New per-model matrix entry for `internlm/Intern-S1-Pro-FP8`.
autotest/configs/internlm/Intern-S1-Pro-BF16.yml	New per-model matrix entry for `internlm/Intern-S1-Pro-BF16`.
autotest/configs/internlm/Intern-S1-mini.yml	New per-model matrix entry for `internlm/Intern-S1-mini`.
autotest/configs/google/gemma-7b-it.yml	New per-model matrix entry for `google/gemma-7b-it`.
autotest/configs/google/gemma-3-27b-it.yml	New per-model matrix entry for `google/gemma-3-27b-it`.
autotest/configs/google/gemma-3-12b-it.yml	New per-model matrix entry for `google/gemma-3-12b-it`.
autotest/configs/google/gemma-2-9b-it.yml	New per-model matrix entry for `google/gemma-2-9b-it`.
autotest/configs/google/gemma-2-27b-it.yml	New per-model matrix entry for `google/gemma-2-27b-it`.
autotest/configs/deepseek-ai/DeepSeek-V3.1.yml	New per-model matrix entry for `deepseek-ai/DeepSeek-V3.1`.
autotest/configs/deepseek-ai/DeepSeek-V2-Lite-Chat.yml	New per-model matrix entry for `deepseek-ai/DeepSeek-V2-Lite-Chat`.
autotest/configs/deepseek-ai/deepseek-moe-16b-chat.yml	New per-model matrix entry for `deepseek-ai/deepseek-moe-16b-chat`.
autotest/configs/codellama/CodeLlama-7b-Instruct-hf.yml	New per-model matrix entry for `codellama/CodeLlama-7b-Instruct-hf`.
autotest/configs/codellama/CodeLlama-7b-hf.yml	New per-model matrix entry for `codellama/CodeLlama-7b-hf`.
autotest/configs/bigcode/starcoder2-7b.yml	New per-model matrix entry for `bigcode/starcoder2-7b`.
autotest/configs/baichuan-inc/Baichuan2-7B-Chat.yml	New per-model matrix entry for `baichuan-inc/Baichuan2-7B-Chat`.
autotest/configs/baichuan-inc/Baichuan2-13B-Chat.yml	New per-model matrix entry for `baichuan-inc/Baichuan2-13B-Chat`.
autotest/config_testascend.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_test.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_legacy.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_h800.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_h_legacy.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_ascend.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_5080.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_5080_legacy.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_3090.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/config_3090_legacy.yml	Removes legacy flat config (superseded by env_paths + per-model matrices).
autotest/benchmark/test_prefixcache_performance.py	Expands prefix-cache performance coverage by attempting to add prefix-caching variants.
.github/workflows/pr_ete_test.yml	Updates checkout action version.
.github/workflows/mllm_api_eval.yml	Updates checkout action version.
.github/workflows/evaluate.yml	Updates checkout action version.
.github/workflows/docker_nightly.yml	Updates checkout action version and extends nightly test image build steps.
.github/workflows/daily_ete_test_5080.yml	Refactors test matrix to use `DEPS_PROFILE` instead of legacy env tags and adjusts install steps.
.github/workflows/daily_ete_test_3090.yml	Refactors test matrix to use `DEPS_PROFILE` instead of legacy env tags and adjusts install steps.
.github/workflows/benchmark.yml	Switches benchmark container tag default to `nightly-test-cu12.8`.
.github/workflows/api_eval.yml	Updates checkout action version.
.github/scripts/eval_base_config.py	Reduces OpenCompass dataset import set and summary output configuration.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

lvhan028 · 2026-05-28T12:11:42Z

    timeout-minutes: 150
+    strategy:
+      matrix:
+        deps_profile: ['', 'transformers==4.57.6']


Can transformers v5.6.0 work?

lvhan028 · 2026-05-28T12:14:16Z

+  quantization:
+    pytorch:
+    - kvint4
+    - kvint8


We can remove its quantization test

lvhan028 · 2026-05-28T12:14:58Z

+  - quantization
+  quantization:
+    turbomind:
+    - awq
+    - kvint4
+    - kvint8


We can remove lagecy models' quantization test

lvhan028 · 2026-05-28T12:26:18Z

          RUN python3 -m pip install --no-cache-dir -r /tmp/requirements/lite.txt && \
              python3 -m pip install --no-cache-dir -r /tmp/requirements/test.txt && \
-              pip install --no-cache-dir ${{ env.OFFLINE_REQUIREMENTS }}
+              pip install --no-cache-dir ${{ env.OFFLINE_REQUIREMENTS }} && \


Could you let us know which pypi package downgrade the "nvidia-nccl-cu12"?

accelerate

#8 2.208 Collecting nvidia-nccl-cu12==2.27.5 (from torch>=2.0.0->accelerate->-r /tmp/requirements/lite.txt (line 1))
#8 2.211 Downloading nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.0 kB)

lvhan028 · 2026-05-28T12:30:10Z

@@ -0,0 +1,18 @@
+# HF model id: deepseek-ai/deepseek-moe-16b-chat


deepseek-ai/deepseek-moe-16b-chat can be eliminated

lvhan028 · 2026-05-28T12:30:57Z

+    tp: 8
+    extra:
+      session-len: 65536


Can we also add dp+ep case?

lvhan028 · 2026-05-28T12:41:18Z

@@ -0,0 +1,30 @@
+# HF model id: internlm/internlm3-8b-instruct-awq


We can remove internlm3 series models

lvhan028 · 2026-05-28T13:13:00Z

+  deps:
+    transformers: 4.57.6


Is "4.57.6" a must?

lvhan028 · 2026-05-28T13:28:33Z

+  - name: pytorch
+    communicators:
+    - nccl


pytorch engine doesn't support cp. Therefore, please remove it

lvhan028 · 2026-05-28T13:32:10Z

+    pytorch:
+    - kvint8


It can be removed safely

lvhan028 · 2026-05-29T11:16:59Z

@@ -0,0 +1,25 @@
+# HF model id: meta-llama/Meta-Llama-3.1-70B-Instruct


Can it be merged with "Meta-Llama-3-1-70B-Instruct.yml"?

lvhan028 · 2026-05-29T11:18:26Z

@@ -0,0 +1,28 @@
+# HF model id: meta-llama/Meta-Llama-3.1-8B-Instruct


Can we merge it into 'Meta-Llama-3-1-8B-Instruct'?

lvhan028 · 2026-05-29T11:21:23Z

@@ -0,0 +1,33 @@
+# HF model id: Qwen/Qwen1.5-MoE-A2.7B-Chat


We can remove this test case

zhulinJulia24 and others added 30 commits April 17, 2026 09:47

update

330ebe4

Merge branch 'InternLM:main' into add_k4v2

82449bb

update

7aa88bf

Merge branch 'InternLM:main' into add_k4v2

e042d89

update

515dba9

Merge branch 'InternLM:main' into add_k4v2

cb8bf8d

Merge branch 'InternLM:main' into add_k4v2

235f05b

Merge branch 'InternLM:main' into add_k4v2

f78d53f

update

bf6aa75

Merge branch 'InternLM:main' into add_k4v2

960c767

update

5aa09c2

update

46a804a

update

c6d277d

update

c29d5ad

Merge branch 'InternLM:main' into add_k4v2

11c0eeb

Merge branch 'InternLM:main' into add_k4v2

32d4b03

Merge branch 'InternLM:main' into add_k4v2

6f169f3

update

d9f2394

update

87273e9

Merge branch 'InternLM:main' into add_k4v2

684a26f

update

963de13

merge main

9537be0

remove qwen3.5 from prefix cache case

df74c36

update

76f70e5

upgrade case style

d86c74f

Merge branch 'InternLM:main' into add_k4v2

3081459

update

bc4a350

Merge upstream/main into add_k4v2

112638c

update

ee6cdc7

Merge branch 'InternLM:main' into add_k4v2

2cd0d9f

zhulinJulia24 added 5 commits May 27, 2026 20:10

update

d649c90

update

a5e88d2

update

a473a1b

update

fbdacb0

update

64b9f93

Copilot AI review requested due to automatic review settings May 28, 2026 05:40

Copilot started reviewing on behalf of zhulinJulia24 May 28, 2026 05:40 View session

update

19fe20a

Copilot AI reviewed May 28, 2026

View reviewed changes

zhulinJulia24 and others added 5 commits May 28, 2026 13:58

Potential fix for pull request finding

5f46267

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

update

27bc946

update

2991cac

update readme

9fa1535

merge main

806b74b

lvhan028 reviewed May 28, 2026

View reviewed changes

Comment thread autotest/configs/bigcode/starcoder2-7b.yml

lvhan028 reviewed May 28, 2026

View reviewed changes

Comment thread autotest/configs/unsloth/gpt-oss-20b-BF16.yml

lvhan028 reviewed May 28, 2026

View reviewed changes

Comment thread autotest/configs/THUDM/glm-4-9b-chat.yml

lvhan028 reviewed May 28, 2026

View reviewed changes

lvhan028 reviewed May 29, 2026

View reviewed changes

		@@ -0,0 +1,18 @@
		# HF model id: deepseek-ai/deepseek-moe-16b-chat

		@@ -0,0 +1,30 @@
		# HF model id: internlm/internlm3-8b-instruct-awq

		@@ -0,0 +1,25 @@
		# HF model id: meta-llama/Meta-Llama-3.1-70B-Instruct

		@@ -0,0 +1,28 @@
		# HF model id: meta-llama/Meta-Llama-3.1-8B-Instruct

Conversation

zhulinJulia24 commented May 28, 2026

Motivation

Modification

BC-breaking (Optional)

Use cases (Optional)

Checklist

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lvhan028 May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lvhan028 May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

lvhan028 May 29, 2026 •

edited

Loading

lvhan028 May 29, 2026 •

edited

Loading