Benchmark
HellaSwag
Scores are percentages. Higher is better.
- Models and systems measured
- 56
- Scores recorded
- 56
- Sources
- 3
- IQ weight
- 0.00
Score distribution
- 76.8%
- Mean
- 81.5%
- Median
- 14.8%
- Std dev
- 55.5%
- 10th percentile
- 89.2%
- 90th percentile
- 33%
- Min
This benchmark has a weight of 0.00 for chat models, measured over 54 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model or system | Score |
|---|---|---|
| 1 | Claude 3 Opus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 95.4% |
| 2 | GPT 4· Epoch AI Benchmarking Hub dataValues from 2 sources
| 95.3% |
| 3 | GPT 4 32k· Epoch AI Benchmarking Hub data | 95.3% |
| 4 | Gemini 1.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 93.3% |
| 5 | MiMo V2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.8% |
| 6 | Llama 3.1 405B· Epoch AI Benchmarking Hub data | 89.2% |
| 7 | Claude 3 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89% |
| 8 | DeepSeek V3· Epoch AI Benchmarking Hub data | 88.9% |
| 9 | Command R Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.6% |
| 10 | Qwen2 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.6% |
| 11 | Mixtral 8X7b V0.1· Epoch AI Benchmarking Hub data | 86.7% |
| 12 | Gemini 1.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.5% |
| 13 | Gemma 2 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.4% |
| 14 | Claude 3 Haiku· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.9% |
| 15 | Llama 3.1 Nemotron 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.6% |
| 16 | Text Davinci 003· Epoch AI Benchmarking Hub data | 85.5% |
| 17 | Qwen2.5 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.2% |
| 18 | Qwen2.5 72B· Epoch AI Benchmarking Hub data | 84.8% |
| 19 | phi-3.5-moe-instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.8% |
| 20 | Mistral Nemo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.5% |
| 21 | Qwen 2.5 Coder 32B· Epoch AI Benchmarking Hub data | 83% |
| 22 | Qwen 2.5 Coder 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83% |
| 23 | Phi-3 Medium 128K Instruct· Epoch AI Benchmarking Hub data | 82.4% |
| 24 | gemma-7b· Epoch AI Benchmarking Hub data | 82.2% |
| 25 | Gemma 2 9B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.9% |
| 26 | Text Davinci 002· Epoch AI Benchmarking Hub data | 81.5% |
| 27 | Mistral 7B v0.1· Epoch AI Benchmarking Hub data | 81% |
| 28 | Llama 2 13b· Epoch AI Benchmarking Hub data | 80.7% |
| 29 | qwen2p5-coder-14b· Epoch AI Benchmarking Hub data | 80.2% |
| 30 | text-davinci-001· Epoch AI Benchmarking Hub data | 79.3% |
| 31 | LLaMA 13B· Epoch AI Benchmarking Hub data | 79.2% |
| 32 | davinci· Epoch AI Benchmarking Hub data | 77.5% |
| 33 | Llama 2 7b· Epoch AI Benchmarking Hub data | 77.2% |
| 34 | Phi-3 Small 8K Instruct· Epoch AI Benchmarking Hub data | 77% |
| 35 | Qwen2.5 Coder 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 76.8% |
| 36 | Qwen2.5-Coder-7B· Epoch AI Benchmarking Hub data | 76.8% |
| 37 | Phi-3 Mini 4K Instruct· Epoch AI Benchmarking Hub data | 76.7% |
| 38 | Yi-6B· Epoch AI Benchmarking Hub data | 74.4% |
| 39 | gemma-2b· Epoch AI Benchmarking Hub data | 71.4% |
| 40 | qwen2p5-coder-3b· Epoch AI Benchmarking Hub data | 70.9% |
| 41 | Llama 3.2 3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 69.8% |
| 42 | Phi 4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 69.1% |
| 43 | curie· Epoch AI Benchmarking Hub data | 68.2% |
| 44 | Text Curie 001· Epoch AI Benchmarking Hub data | 67.6% |
| 45 | qwen2p5-coder-1p5b· Epoch AI Benchmarking Hub data | 61.8% |
| 46 | ChatGLM2 6B· Epoch AI Benchmarking Hub data | 57% |
| 47 | Text Babbage 001· Epoch AI Benchmarking Hub data | 56.1% |
| 48 | babbage· Epoch AI Benchmarking Hub data | 55.5% |
| 49 | phi-2· Epoch AI Benchmarking Hub data | 53.6% |
| 50 | qwen2p5-coder-0p5b· Epoch AI Benchmarking Hub data | 48.4% |
Showing 1–50 of 54 results.
Not ranked
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| DeepSeek V4.1 Flash BaseEM · 0-shot· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87.2% |
| MiMo V2 Flash Baseofficial Base table · 10-shot· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 88.5% |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.