Benchmark Leaderboard

Top models by HellaSwag score

Sorts are only valid inside this benchmark. A score gap of this benchmark does not map to the same gap on another benchmark, so validate any short-list by checking peer rows and model context/price fit.

GPQA HellaSwag HumanEval MMLU

#	Model	Score	Version	Source
1	Clef-omni	99.1	decision-format (Decision Index 0.2.1 internal run)	https://huggingface.co/Cloudflare/clef-omni
2	Clef-flash	98.6	decision-format (Decision Index 0.2.1 internal run)	https://huggingface.co/Cloudflare/clef-flash
3	Clef	98.2	decision-format (Decision Index 0.2.1 internal run)	https://huggingface.co/Cloudflare/clef
4	GPT-4o (05-13)	96.4	10-shot	https://crfm.stanford.edu/helm/classic/latest/
5	Claude 3.5 Sonnet	96.2	10-shot	https://crfm.stanford.edu/helm/classic/latest/
6	Llama 3.1 405B	95.8	10-shot	https://ai.meta.com/blog/meta-llama-3-1/
7	DeepSeek V3	95.7	10-shot	https://arxiv.org/abs/2412.19437
8	Qwen2.5-72B-Instruct	95.6	standard	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
9	Llama 3.1 70B Instruct	94.2	10-shot	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
10	Mistral Medium	93.9	10-shot	research
11	Mistral Large 2	93.8	10-shot	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
12	Mixtral 8x22B v0.1	93.8	10-shot	research
13	Falcon 180B	92.7	10-shot	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
14	Gemma 2 27B	92.6	10-shot	research
15	Llama 3 70B	92.4	10-shot	research
16	Qwen2-7B	92	10-shot	https://arxiv.org/abs/2407.10671
17	Mistral NeMo Instruct (2407)	91.8	10-shot	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
18	StarCoder2 15B	91.7	10-shot	research
19	DeepSeek Coder V2 Lite	91.4	10-shot	https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
20	Llama 3 8B Instruct	91.1	10-shot	research

Interpretation

Trust when:

Use top rows for shortlisting, then validate provider coverage and feature fit before selecting a model.
Cross-benchmark consistency is a stronger signal than one-off score dominance.

Don't trust when:

Don't compare raw score gaps across different benchmark scales as equivalent.
Missing rows in the heatmap can hide benchmark blind spots.
No trend data means this is a snapshot; expect drift in fresh refresh cycles.

Cross-benchmark heatmap

Compare top models across selected benchmark families to spot strengths and weak spots. On narrow screens, scroll horizontally to inspect every benchmark.

Model	GPQA	HellaSwag	HumanEval	MMLU
Clef-omni	47.4	99.1	—	92.7
Clef-flash	51.0	98.6	—	91.8
Clef	48.0	98.2	—	90.3
GPT-4o (05-13)	—	96.4	90.2	88.7
Claude 3.5 Sonnet	—	96.2	92.0	88.7
Llama 3.1 405B	51.5	95.8	89.0	88.6