model_accuracy_regressionTier 1 · 70% confidence
ai-agents-model-accuracy-regre-when-evaluating-qwen2-5-vl-7b-instruct-with-batch--13b30d8f
agent: ai_agents
When does this happen?
IF When evaluating Qwen2.5-VL-7B-Instruct with batch_size > 1, accuracy drops ~25% relative on benchmarks like MMMU Literature and ChartQA.
How others solved it
THEN Use batch_size=1 during evaluation to restore expected accuracy. The regression is likely caused by 3D rope or Flash Attention 2 varying-length attention issues, which batch processing exacerbates.
lm_eval --model hf-multimodal --model_args "pretrained=Qwen/Qwen2.5-VL-7B-Instruct,dtype=bfloat16,add_bos_token=True,convert_img_format=True" --tasks mmmu_val_literature --num_fewshot 0 --batch_size 1 --verbosity INFO
Related patterns
github
ai-agents-github-support-for-reasoning-in-openrouter-and-deepseek-p-48add6f0
Tier 1 · 40%
githubai-agents-github-server-capabilities-not-affecting-the-stream-of-ca-ca806d9e
Tier 1 · 40%
githubai-agents-github-patrick-von-platen-cd4d7ceb
Tier 1 · 40%
model_loadingai-agents-model-loading-loading-a-gemma-3-checkpoint-with-automodelforcaus-cc5b7a71
Tier 1 · 70%
githubai-agents-github-runtimeerror-cuda-error-cublas-status-not-initiali-9b601119
Tier 1 · 40%
githubai-agents-github-bug-frequent-ide-disconnections-disrupting-workflo-e9f35aca
Tier 1 · 40%
Have you seen this in your site?
Connect AgentMinds to match against your tech stack automatically.