Mistral AIDevstral Small 2505
Devstral Small 2505 is a Mistral AI open-weights model, released 21 May 2025. Its best configuration ranks #620 of 785 on the BenchLeader Index at 41.7 ±1.5, in the lower half. The ± is the point: 278 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in long context (40) and lowest in instruction following (35). At $0.150 per million tokens blended it is among the cheapest fifth of ranked models. It has been superseded by Devstral Small 2507, which is the version to compare against today. Last measured 21 May 2025.stale: no new result in six months
- Blended price
- $0.150/M
- $0.100 in · $0.300 out
- Output speed
- –
- First answer
- –
- Context
- 256k
- Full answer
- –
- Cost per run
- –
- Released
- 21 May 2025
- Overall index42
- Reasoning39
- Agents & tools39
- Knowledge36
- Instruction following35
- Long context40
- Composite39
Versions
Mistral AI has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Devstral Small 2 | 9 Dec 2025 | 39.5 | #683 |
| Devstral Small 2507 | 10 Jul 2025 | 41.9 | #613 |
| Devstral Small 2505this page | 21 May 2025 | 41.7 | #620 |
How its index has moved
9 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| GPQA Diamond (AA)not in index | 43.4%#457 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 4.0%#548 | Artificial Analysis | |
| Kagi LLM Benchmark | 37.7%#112 | Kagi LLM Benchmark | |
| CritPt | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| SWE-bench Verified (any scaffold)not in index | 46.8%#39 | SWE-bench |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 6.1%#266 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 38.0%#225 | Artificial Analysis |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -56.8#441 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 15.9%#402 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 13.7%#357 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 31.6%#349 | Artificial Analysis |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 32.0%#387 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 8.7#428 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | – | – |
| Summarize a 30-page report | 12,000 / 600 | $0.0014 | – | – |
| Code edit | 6,000 / 1,500 | $0.0011 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.0072 | – | – |
| Structured extraction | 2,000 / 200 | $0.0003 | – | – |
More from Mistral AI
See also
Questions people ask
- What are Devstral Small 2505's benchmark scores?
- As of 25 Oct 2026, Devstral Small 2505 scores 41.7 ±1.5 on the BenchLeader Index, strongest in long context (40) and weakest in instruction following (35). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does Devstral Small 2505 rank?
- Devstral Small 2505 is #620 of 785 ranked configurations on 25 Oct 2026. 278 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Devstral Small 2505 cost per 1M tokens?
- As of 25 Oct 2026, Devstral Small 2505 costs $0.100 per 1M input tokens and $0.300 per 1M output tokens, $0.150 blended (three input tokens per output token).
- When was Devstral Small 2505 released?
- Mistral AI released Devstral Small 2505 on 21 May 2025.
Data as of 25 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Devstral Small 2505: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/devstral-small.