Claude Sonnet 4.5 Vision on HumanEval: score, rank and what it means
Claude Sonnet 4.5 Vision scores 87.0 on HumanEval, placing it #87 of 600 models in the index. The benchmark average is 74.0 and the leader is GPT-5.2 Pro at 96.8.
Score distribution
The chart shows the top 10 models on HumanEval, with Claude Sonnet 4.5 Vision highlighted. Longer bars mean higher scores.
Top 10 leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 2 | Gemini 3 Pro Pro | Google DeepMind | 200000 | $19.80 | 87.3 |
| 3 | o4 Max | OpenAI | 2000000 | $19.51 | 86.2 |
| 4 | Grok 4 Reasoning | xAI | 2000000 | $19.67 | 87.8 |
| 5 | Claude Opus 4.5 | Anthropic | 400000 | $19.42 | 87.7 |
| 6 | GPT-5.2 Max | OpenAI | 1000000 | $21.39 | 89.2 |
| 7 | Claude Opus 4.5 Pro | Anthropic | 200000 | $20.98 | 89.1 |
| 8 | Grok 4 Max | xAI | 2000000 | $19.70 | 86.6 |
| 9 | GPT-5.2 Turbo | OpenAI | 1000000 | $20.73 | 87.4 |
| 10 | GPT-5.1 Reasoning | OpenAI | 400000 | $20.27 | 88.3 |
What this score means
On HumanEval, Claude Sonnet 4.5 Vision sits at #87 of 600. The leader, GPT-5.2 Pro, scores 96.8, so the gap to the top is 9.8 points. The median model in this index scores 77.4, which puts Claude Sonnet 4.5 Vision above the midpoint.
Anthropic models average 86.0 on this benchmark. Claude Sonnet 4.5 Vision is 1.0 points above that provider line, so it is a stronger result than the average Anthropic model.
Among frontier models, the average is 90.8. Claude Sonnet 4.5 Vision is 3.8 points below that group. This is the more useful comparison if you are shortlisting within the same tier rather than across the whole index.
Read a 87.0 as a relative signal, not an absolute guarantee. Benchmark scores depend on prompt format, evaluation harness and contamination controls. Use this page to narrow the shortlist, then test the model on your own data.
Price context
At $16.99 per million input tokens, Claude Sonnet 4.5 Vision is above the median input price of $4.00 across the 600 models scored on HumanEval. The models immediately around it in the ranking are shown below so you can compare score and price together.
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Llama 4 Maverick | Meta AI | 200000 | $5.36 | 81.6 |
| Kimi K1.5 | Moonshot AI | 128000 | $5.19 | 80.7 |
| Kimi K2 | Moonshot AI | 128000 | $5.85 | 83.0 |
| Claude Haiku 4.5 Turbo | Anthropic | 200000 | $14.37 | 80.0 |
| Claude Sonnet 4.5 Vision | Anthropic | 128000 | $16.99 | 83.8 |
| Gemini 2.5 Pro Max | Google DeepMind | 200000 | $16.94 | 84.8 |
| Qwen3 | Alibaba Cloud | 128000 | $5.04 | 80.3 |
Other benchmarks for Claude Sonnet 4.5 Vision
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 84.8 | 66 | 89th | |
| GPQA Diamond | 79.6 | 64 | 89th | |
| SWE-bench Verified | 73.7 | 58 | 90th | |
| MATH-500 | 90.6 | 38 | 94th | |
| AIME 2025 | 86.4 | 24 | 96th | |
| LiveCodeBench | 76.5 | 74 | 88th | |
| IFEval | 89.9 | 74 | 88th | |
| TruthfulQA | 83.7 | 88 | 85th | |
| HellaSwag | 89.4 | 101 | 83th | |
| MMMU | 79.7 | 60 | 90th | |
| Arena Elo | 1447.6 | 81 | 87th |
FAQ
What score does Claude Sonnet 4.5 Vision get on HumanEval?
Claude Sonnet 4.5 Vision scores 87.0 on HumanEval in this index. The benchmark average is 74.0 and the top score is 96.8.
How does Claude Sonnet 4.5 Vision rank on HumanEval?
It ranks #87 of 600 models, in the 86th percentile. The leader is GPT-5.2 Pro at 96.8.
Is Claude Sonnet 4.5 Vision good at HumanEval?
Yes, relative to the index: it is above the median and inside the top half of scores. The result should be read alongside its price and context window.
How is the HumanEval score calculated?
HumanEval is a coding benchmark. Scores in this index are modelled estimates on a 0–100 scale (or Elo for Arena), derived from capability tier and benchmark difficulty. They are not official leaderboard results.
What does Claude Sonnet 4.5 Vision cost?
Claude Sonnet 4.5 Vision costs $16.99 per million input tokens and $82.33 per million output tokens.