BenchLeader

Devstral Small 2507

As of 16 Nov 2026, Devstral Small 2507 by Mistral AI is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories. API pricing: $0.150 per 1M tokens blended (3 input : 1 output), list price on 16 Nov 2026. Released 10 Jul 2025.

Key facts (as of 16 Nov 2026)

  • Devstral Small 2507 API pricing: $0.150 per 1M tokens blended (3 input : 1 output) (list price, 16 Nov 2026).
  • Devstral Small 2507 was released on 10 Jul 2025 (Mistral AI).

In detail. Devstral Small 2507 is a Mistral AI open-weights model, released 10 Jul 2025. It is not yet ranked: the index needs at least five independent results across two categories. Last measured 10 Jul 2025.stale: no new result in six months

Blended price
–
Output speed
–
First answer
–
Context
256k
via Artificial Analysis
Cost per task
–
Tokens per task
–
Released
10 Jul 2025
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Agents & tools39
  2. Instruction following38
  3. Long context33

Get an email when Devstral Small 2507 gets its first benchmark results

And when Mistral AI ships a new model, plus the weekly review on Mondays.

Choose providers and emails

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.

ModelReleasedIndexRank
Devstral Small 29 Dec 202539.4#544
Devstral Small 2507this page10 Jul 2025––
Devstral Small 250521 May 2025––

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.

Metadata

Benchmarknot statedSourceTrend
AA Openness Indexnot in index44.4#97Artificial Analysis

Reasoning

Benchmarknot statedSourceTrend
GPQA Diamond (AA)not in index41.4%#480Artificial Analysis
Humanity's Last Exam (AA)not in index3.8%#594Artificial Analysis
CritPtnot in index0.0%#337Artificial Analysis

Coding

Benchmarknot statedSourceTrend
SWE-bench Verified (any scaffold)not in index38.0%#49SWE-bench
Design Arena Websitenot in index794#191Design Arena
LiveCodeBench (AA)not in index25.4%#227Artificial Analysis

Agents & tools

Benchmarknot statedSourceTrend
Terminal-Bench Hard6.1%#271Artificial Analysis
τ²-Bench Telecom (AA)not in index28.4%#272Artificial Analysis

Maths

Benchmarknot statedSourceTrend
AIME 2025 (AA)not in index29.3%#178Artificial Analysis

Instruction following

Benchmarknot statedSourceTrend
IFBench34.6%#321Artificial Analysis

Long context

Benchmarknot statedSourceTrend
AA-LCR v1.118.7%#465Artificial Analysis

Composite

Benchmarknot statedSourceTrend
AA Intelligence Index v4.3.2not in index7.6#475Artificial Analysis
AA Progress Index v1not in index153.1#554Artificial Analysis

More from Mistral AI

New this month

See also

Questions people ask

Why does Devstral Small 2507 have no BenchLeader Index score?
The index needs at least five independent results across at least two categories, and as of 16 Nov 2026 Devstral Small 2507 does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
When was Devstral Small 2507 released?
Mistral AI released Devstral Small 2507 on 10 Jul 2025.

Data as of 16 Nov 2026. Compare with another model.

Cite as: BenchLeader, “Devstral Small 2507: benchmarks, pricing, speed and rank”, data as of 16 Nov 2026, https://www.benchleader.com/models/devstral-small-2507.