Workflow runs · lm-sys/FastChat · GitHub

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows

98 workflow runs

98 workflow runs

fix(openai_api_server): use exclude_none for completion stream finish chunks Python package #4065: Pull request #3872 opened by Chessing234

40s fix-openai-api-completion-stream-finish-exclude-none

fix-openai-api-completion-stream-finish-exclude-none

40s

Fix create_completion capping all batch prompts to the shortest context's max_tokens Python package #4064: Pull request #3867 opened by Chessing234

54s fix/completion-max-tokens-shared-across-prompts

fix/completion-max-tokens-shared-across-prompts

54s

fix(arena): pass full conversation history to moderate_input in vision anony arena Python package #4062: Pull request #3864 opened by Chessing234

43s fix/vision-arena-moderate-full-conversation

fix/vision-arena-moderate-full-conversation

43s

fix: add weights_only=True to torch.load calls in apply_delta.py Python package #4061: Pull request #3863 opened by Chessing234

40s fix/apply-delta-weights-only-torch-load

fix/apply-delta-weights-only-torch-load

40s

fix: handle missing pad_token for Llama-3 and tokenizers without unk_… Python package #4060: Commit 36ae08f pushed by infwinston

42s claude/fix-critical-issue-L2emR

claude/fix-critical-issue-L2emR

42s

fix: guard vote_last_response against None state Python package #4059: Pull request #3862 opened by Chessing234

39s fix/vote-null-state-check

fix/vote-null-state-check

39s

Fix load_demo_single called with list instead of Context in gradio_web_server Python package #4058: Pull request #3860 opened by Chessing234

42s fix/load-demo-single-context

fix/load-demo-single-context

42s

Fix bias='lora_only' state-dict collection in train_lora Python package #4057: Pull request #3859 opened by Chessing234

46s fix/train-lora-bias-only-state-dict

fix/train-lora-bias-only-state-dict

46s

Fix load_log_files UnboundLocalError when all FileNotFoundError retries fail Python package #4056: Pull request #3858 opened by Chessing234

45s fix/load-log-files-unbound-lines-after-retries

fix/load-log-files-unbound-lines-after-retries

45s

Fix load_compress_model retry using bitwise ~ on bool instead of logical not Python package #4055: Pull request #3857 opened by Chessing234

33s fix/compression-use-fast-bitwise-not

fix/compression-use-fast-bitwise-not

33s

Fix ModelPermission.is_blocking type annotation (str -> bool) Python package #4054: Pull request #3856 opened by Chessing234

40s fix/model-permission-is-blocking-type

fix/model-permission-is-blocking-type

40s

Fix Monitor.update_model_call_limit always rejecting every call Python package #4053: Pull request #3855 opened by Chessing234

35s fix/call-monitor-update-limit-unreachable

fix/call-monitor-update-limit-unreachable

35s

Return after no-worker yield in Controller.worker_api_generate_stream Python package #4052: Pull request #3854 opened by Chessing234

38s fix/controller-generate-stream-missing-return

fix/controller-generate-stream-missing-return

38s

Fix VLLMWorker.generate_stream referencing module-level engine instead of self Python package #4051: Pull request #3853 opened by Chessing234

34s fix/vllm-worker-engine-self-reference

fix/vllm-worker-engine-self-reference

34s

Fix crash in vote_last_response when state is None after clear_history Python package #4050: Pull request #3852 opened by Chessing234

35s fix/vote-last-response-null-state

fix/vote-last-response-null-state

35s

Fix add_text/regenerate return tuple size mismatch in single-model server Python package #4049: Pull request #3851 opened by Chessing234

34s fix/issue-3783-return-tuple-size

fix/issue-3783-return-tuple-size

34s

Fix moderation only checking left model's conversation in arena battles Python package #4048: Pull request #3850 opened by Chessing234

39s fix/arena-moderation-wrong-state-index

fix/arena-moderation-wrong-state-index

39s

fix: use identity check for None comparison (PEP 8 E711) Python package #4047: Pull request #3849 opened by harshadkhetpal

37s harshadkhetpal:fix/pep8-is-none-model-worker

harshadkhetpal:fix/pep8-is-none-model-worker

37s

Fix off-by-one in is_partial_stop causing false positives Python package #4046: Pull request #3848 opened by Chessing234

41s fix/is-partial-stop-off-by-one

fix/is-partial-stop-off-by-one

41s

Fix ModelPermission.is_blocking type annotation (str -> bool) Python package #4045: Pull request #3846 opened by Chessing234

36s fix/is-blocking-type

fix/is-blocking-type

36s

Fix bitwise NOT (~) on boolean in tokenizer fallback Python package #4044: Pull request #3845 opened by Chessing234

34s fix/bitwise-not-use-fast

fix/bitwise-not-use-fast

34s

Fix off-by-one in completion_tokens count in generate_stream Python package #4043: Pull request #3843 opened by Chessing234

43s fix/completion-tokens-off-by-one

fix/completion-tokens-off-by-one

43s

Fix moderation filter checking wrong model in arena side-by-side views Python package #4042: Pull request #3842 opened by Chessing234

44s fix/moderation-filter-arena-bypass

fix/moderation-filter-arena-bypass

44s

Fix typos in judge prompts: "You evaluation" → "Your evaluation" Python package #4041: Pull request #3841 opened by Chessing234

38s fix/issue-2802-judge-prompt-typos

fix/issue-2802-judge-prompt-typos

38s

Fix batch embedding averaging for batch_size > 1 Python package #4040: Pull request #3839 opened by Chessing234

38s fix/issue-3785-batch-embedding-averaging

fix/issue-3785-batch-embedding-averaging

38s