Mistral Small 4
Mistral Small 4Alternatives?
CheaperGemma 4 31BthinkingOpen+57%-39%$0.10 / $0.34Context: 262K → 256K
BetterGPT OSS 20BhighOpen+35%-38%$0.011Context: 262K → 131K+2
OpenNemotron Cascade 2 30B A3B+67%——No image inputCheaperGemma 4 12BthinkingOpen+53%-43%$0.10 / $0.30Context: 262K → 256K+1CheaperGPT-5 Nanohigh+40%-48%$0.05 / $0.40Gemma 4 31BthinkingOpen+57%-39%$0.10 / $0.34Context: 262K → 256KGemma 4 12BthinkingOpen+53%-43%$0.10 / $0.30Context: 262K → 256K+1GPT-5 Nanohigh+40%-48%$0.05 / $0.40GPT OSS 20BhighOpen+35%-38%$0.011Context: 262K → 131K+2
GLM 4.7 FlashthinkingOpen+26%-45%$0.061 / $0.40Context: 262K → 200K+2GPT OSS 20BhighOpen+35%-38%$0.011Context: 262K → 131K+2Nemotron 3 Nano 30B A3BthinkingOpen+48%same$0.017Context: 262K → 131K+2
MiMo v2.5thinkingOpen+39%+144%$0.042Memory (4-bit): 72 GB → 186 GB+1GPT-5 Minihigh+57%+212%$0.053
Solar Pro 3high+48%+240%$0.058Context: 262K → 131K+2Nemotron Cascade 2 30B A3BOpen+67%——No image input
MiniMax M2.7thinkingOpen+57%+521%$0.11Context: 262K → 205K+3
Qwen 3.5 122B A10BthinkingOpen+57%+1,794%$0.32Memory (4-bit): 72 GB → 75 GB+1Gemma 4 31BthinkingOpen+57%-39%$0.10 / $0.34Context: 262K → 256KQwen 3.5 27BthinkingOpen+57%+214%$0.30 / $2.40No PDF inputMistral Medium 3.5highOpen+43%+2,400%$0.43Memory (4-bit): 72 GB → 77 GBMistral Large 3non-thinkingOpen-25%+82%$0.031Memory (4-bit): not verifiedMinistral 3 14B 2512non-thinkingOpen-33%+13%$0.019Devstral Mediumnon-thinkingOpen-38%—$0.40 / $2Context: 262K → 128K+3Ministral 3 8B 2512Open-40%-35%$0.011Ministral 3 3B 2512Open-44%-54%$0.008Context: 262K → 131KMistral Small 3non-thinkingOpen-45%—$0.05 / $0.08Context: 262K → 32K+2Magistral Medium 1thinking-48%——Context: 262K → 40K+1All Mistral models →Run it yourself
estimateHistory
3 versionsBenchmarks
in Chat & KnowledgeFAQ
How much does Mistral Small 4 cost?›
$0.15 per 1M input tokens and $0.60 per 1M output tokens; cached input $0.015 per 1M. A typical benchmark task costs about $0.017 at default effort.
Is Mistral Small 4 free to use?›
There is no free API; the cheapest paid price is $0.15 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Mistral Small 4, and what is it best at?›
#195 of 336 in Chat & Knowledge, #77 of 134 in Coding, #114 of 135 in Agents & Tool Use, #74 of 96 in Writing & Language, #102 of 125 in Vision & Documents.
Is Mistral Small 4 better than Gemini 2.5 Pro?›
On IFBench (instructions), Mistral Small 4 scores 48.2 vs 48.7 — about the same. It is 95% cheaper per task.
How can I use Mistral Small 4?›
Through the API of Mistral.
What is the context window of Mistral Small 4?›
262K tokens, with up to 210K output tokens.
When was Mistral Small 4 released? Is there a newer version?›
Released 2026-03-16. It is the current version in its line.
Is Mistral Small 4 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Mistral Small 4 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Mistral Small 4 read images?›
It accepts text and images, PDF files as input.
What is Mistral Small 4?›
Mistral Small 4 is an open-weights language model from Mistral, released Mar 2026, the successor to Saba.
What are the best alternatives to Mistral Small 4?›
Gemma 4 31B (cheaper), GPT OSS 20B (better), Nemotron Cascade 2 30B A3B (open weights).