Mistral Small 2506
Mistral Small 2506Compare inon
Benchmarks of Mistral Small 250615
BenchmarkByScoreRank
Legacy · saturated or no longer reported2
No benchmark matches
Ranking?
Retired models are left out of the current rankings.
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-5 Nano (2025-08-07)unstated+38%-8%$0.05 / $0.40
BetterClaude Opus 4.5unstated×2.1+6,567%$5 / $25
OpenKimi K2 0711unstated+59%+483%$0.50 / $2No image input
CheaperQwen3 30B A3B Instruct 2507unstatedOpen+11%-5%$0.09 / $0.30No image input
CheaperCommand R7b-14%-56%$0.037 / $0.15Context: not verified+1ModelScoreCostPer taskWhat changes
GPT-5 Nano (2025-08-07)unstated+38%-8%$0.05 / $0.40Qwen3 30B A3B Instruct 2507unstatedOpen+11%-5%$0.09 / $0.30No image inputCommand R7b-14%-56%$0.037 / $0.15Context: not verified+1
Gemma 3 12BunstatedOpen-18%-58%$0.05 / $0.10Gemma 3 27BunstatedOpen-21%-33%$0.08 / $0.16ModelScoreCostPer taskWhat changes
Claude Opus 4.5unstated×2.1+6,567%$5 / $25Claude Sonnet 4.5unstated+97%+3,900%$3 / $15Claude Haiku 4.5unstated+85%+1,233%$1 / $5o3 (2025-04-16)unstated+70%+2,233%$2 / $8Kimi K2 0711unstatedOpen+59%+483%$0.50 / $2No image inputModelScoreCostPer taskWhat changes
Kimi K2 0711unstatedOpen+59%+483%$0.50 / $2No image inputQwen3 235B A22B Instruct 2507unstatedOpen+40%+37%$0.09 / $0.55No image inputNNanbeige4 3B Thinking 2511Open+38%——Context: 128K → 66K+1Qwen3 32BunstatedOpen+31%+717%$0.70 / $2.80Context: 128K → 33K+1Command AunstatedOpen+25%+2,817%$2.50 / $10No image inputModelScoreCostPer taskWhat changes
Mistral Large 4thinkingn/a—$1.61Infinite whitespace stream+4Mistral Medium 3.5highOpenn/a—$0.43Mistral Small 4thinkingOpenn/a—$0.017Mistral Smalln/a—$1 / $3Context: 128K → 32KMinistral 3 14B 2512non-thinkingOpenn/a—$0.019Ministral 3 3B 2512Openn/a—$0.008Ministral 3 8B 2512Openn/a—$0.011Mistral Large 3non-thinkingOpenn/a—$0.031All Mistral models →Benchmarks
all categoriesChat & Knowledge
Arena Text · Chinese1,340.4
Arena Text · English1,357.1
Arena Text · French1,336.6
Arena Text · German1,340.2
Arena Text · Japanese1,275.3
Arena Text · Korean1,257.7
Arena Text · non-English1,315.1
Arena Text · Polish1,316.8
Arena Text · Russian1,323
Arena Text · Spanish1,346.9
Agents & Tool Use
BFCL v437.2
Writing & Language
FAQ
How much does Mistral Small 2506 cost?›
$0.10 per 1M input tokens and $0.30 per 1M output tokens (cheapest host: DevPass (LLM Gateway)).
Is Mistral Small 2506 free to use?›
There is no free API; the cheapest paid price is $0.10 per 1M input tokens.
How can I use Mistral Small 2506?›
Through the Mistral API.
What is the context window of Mistral Small 2506?›
128K tokens.
When was Mistral Small 2506 released? Is there a newer version?›
It has been superseded by Mistral Small 4.
Can Mistral Small 2506 read images?›
It accepts text and images as input.
What is Mistral Small 2506?›
Mistral Small 2506 is a language model from Mistral.
What are the best alternatives to Mistral Small 2506?›
GPT-5 Nano (2025-08-07) (cheaper), Claude Opus 4.5 (better), Kimi K2 0711 (open weights).