Codestral 8B on LiveCodeBench: score, rank and what it means
Codestral 8B scores 49.5 on LiveCodeBench, placing it #453 of 600 models in the index. The benchmark average is 61.0 and the leader is Claude Sonnet 4.5 Reasoning at 85.1.
Score distribution
The chart shows the top 10 models on LiveCodeBench, with Codestral 8B highlighted. Longer bars mean higher scores.
Top 10 leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 Reasoning | Anthropic | 400000 | $19.48 | 87.7 |
| 2 | Grok 4 Reasoning | xAI | 2000000 | $19.67 | 87.8 |
| 3 | Claude Opus 4.5 | Anthropic | 400000 | $19.42 | 87.7 |
| 4 | GPT-5.2 Max | OpenAI | 1000000 | $21.39 | 89.2 |
| 5 | Grok 4 Turbo | xAI | 1000000 | $19.21 | 86.7 |
| 6 | GPT-5.1 Turbo | OpenAI | 2000000 | $19.19 | 86.1 |
| 7 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 8 | GPT-5.2 Reasoning | OpenAI | 1000000 | $20.98 | 87.5 |
| 9 | Claude Opus 4.5 Pro | Anthropic | 200000 | $20.98 | 89.1 |
| 10 | Gemini 3 Pro Max | Google DeepMind | 2000000 | $20.57 | 87.9 |
What this score means
On LiveCodeBench, Codestral 8B sits at #453 of 600. The leader, Claude Sonnet 4.5 Reasoning, scores 85.1, so the gap to the top is 35.6 points. The median model in this index scores 64.7, which puts Codestral 8B below the midpoint.
Mistral AI models average 58.1 on this benchmark. Codestral 8B is 8.6 points below that provider line, so it is a weaker result than the average Mistral AI model.
Among small open-weight models, the average is 41.2. Codestral 8B is 8.3 points above that group. This is the more useful comparison if you are shortlisting within the same tier rather than across the whole index.
Read a 49.5 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 $1.18 per million input tokens, Codestral 8B is below the median input price of $4.00 across the 600 models scored on LiveCodeBench. 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 |
|---|---|---|---|---|
| Zephyr Instruct | Hugging Face | 4096 | $1.66 | 62.4 |
| Nova Lite Nano | Amazon | 8192 | $3.29 | 59.5 |
| SmolLM3 | Hugging Face | 8192 | $1.71 | 63.1 |
| Llama 4 Scout 8B | Meta AI | 8192 | $1.46 | 61.2 |
| Codestral 8B | Mistral AI | 4096 | $1.18 | 59.0 |
| GLM-4.5 7B | Zhipu AI | 8192 | $1.32 | 60.2 |
| GPT-NeoX Base | EleutherAI | 16384 | $1.22 | 59.7 |
Other benchmarks for Codestral 8B
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 65.7 | 441 | 27th | |
| GPQA Diamond | 53.3 | 488 | 19th | |
| SWE-bench Verified | 46.1 | 451 | 25th | |
| HumanEval | 62.8 | 459 | 24th | |
| MATH-500 | 57.9 | 492 | 18th | |
| AIME 2025 | 49.9 | 461 | 23th | |
| IFEval | 70.4 | 451 | 25th | |
| TruthfulQA | 65.4 | 455 | 24th | |
| HellaSwag | 72.0 | 495 | 18th | |
| MMMU | 55.6 | 491 | 18th | |
| Arena Elo | 1253.3 | 486 | 19th |
FAQ
What score does Codestral 8B get on LiveCodeBench?
Codestral 8B scores 49.5 on LiveCodeBench in this index. The benchmark average is 61.0 and the top score is 85.1.
How does Codestral 8B rank on LiveCodeBench?
It ranks #453 of 600 models, in the 25th percentile. The leader is Claude Sonnet 4.5 Reasoning at 85.1.
Is Codestral 8B good at LiveCodeBench?
It is below the median for this benchmark, so test it carefully before relying on it for this capability. The result should be read alongside its price and context window.
How is the LiveCodeBench score calculated?
LiveCodeBench 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 Codestral 8B cost?
Codestral 8B costs $1.18 per million input tokens and $5.35 per million output tokens.