Mistral Medium 3.5
Mistral Medium 3.5Alternatives?
CheaperGPT-6.1 Solxhigh×3.8-22%$0.39Remove unsupported parameters+5BetterGPT-6 Lunalow+47%-99%$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 Flashthinking×2.8-26%$0.37Memory (4-bit): 77 GB → 108 GB+6
CheaperMiMo v2.6 ProthinkingOpen×3.6-73%$0.13Memory (4-bit): 77 GB → 615 GB+5
CheaperMiniMax M3.1 Flashthinking×3.5-56%$0.22No PDF input+1GPT-6.1 Solxhigh×3.8-22%$0.39Remove unsupported parameters+5MiMo v2.6 ProthinkingOpen×3.6-73%$0.13Memory (4-bit): 77 GB → 615 GB+5MiniMax M3.1 Flashthinking×3.5-56%$0.22No PDF input+1
Claude Haiku 5.5max×3.2-58%$0.21Remove sampling parameters+5GPT-6 Solhigh×3.2-26%$0.37GPT-6 Lunalow+47%-99%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+44%-98%$0.01MiMo v2.5thinkingOpen+98%-96%$0.019Memory (4-bit): 77 GB → 186 GB+1Claude Haiku 5.5low+96%-95%$0.024Remove sampling parameters+5
Mercury 2high+25%-92%$0.039Context: 262K → 128K+2Qwen3.8 FlashthinkingOpen×2.8-26%$0.37Memory (4-bit): 77 GB → 108 GB+6MMotif 3Open×2.7-71%$0.50 / $2Memory (4-bit): 77 GB → 189 GB+2MiMo v2.6 FlashthinkingOpen×2.5-88%$0.062Memory (4-bit): 77 GB → 186 GB+5Qwen3.8 27BxhighOpen×2.5+101%$1.01max_tokens range · QwenCloud+4
Hy3thinkingOpen×2.4-86%$0.072Context: 262K → 256K+3Mistral Large 4thinking×2.5+126%$1.13Infinite whitespace stream+4Mistral Small 4thinkingOpen-28%-97%$0.015Magistral Medium 1thinking-29%——Context: 262K → 40K+1Magistral Small 1thinkingOpen-45%—$0.70 / $1.16Context: 262K → 128K+2Ministral 3 3B 2512Open-61%-98%$0.008Context: 262K → 131KMinistral 3 14B 2512non-thinkingOpen-66%-96%$0.019Ministral 3 8B 2512Open-69%-98%$0.011Mistral Large 3non-thinkingOpen-70%-94%$0.031Memory (4-bit): not verifiedAll Mistral models →Run it yourself
estimateHistory
3 versionsBenchmarks
in Chat & KnowledgeFAQ
How much does Mistral Medium 3.5 cost?›
$1.50 per 1M input tokens and $7.50 per 1M output tokens; cached input $0.15 per 1M; cheapest via Featherless at $1 / $4. A typical benchmark task costs about $0.50 at high effort.
Is Mistral Medium 3.5 free to use?›
There is no free API; the cheapest paid price is $1 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Mistral Medium 3.5, and what is it best at?›
#151 of 334 in Chat & Knowledge, #75 of 132 in Coding, #80 of 133 in Agents & Tool Use, #69 of 195 in Writing & Language, #81 of 124 in Vision & Documents. Strongest at: Tool calling, Instructions.
Is Mistral Medium 3.5 better than Devstral 2 2512?›
On Humanity's Last Exam, Mistral Medium 3.5 scores 13.8 vs 3.6 — higher.
How can I use Mistral Medium 3.5?›
Through the API of Featherless, Mistral.
What is the context window of Mistral Medium 3.5?›
262K tokens, with up to 210K output tokens.
When was Mistral Medium 3.5 released? Is there a newer version?›
Released 2026-04-29. It is the current version in its line.
Is Mistral Medium 3.5 open source?›
Yes, the weights are open.
Can I run Mistral Medium 3.5 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Mistral Medium 3.5 read images?›
It accepts text and images, PDF files as input.
What is Mistral Medium 3.5?›
Mistral Medium 3.5 is an open-weights language model from Mistral, released Apr 2026, the successor to Devstral 2 2512.
What are the best alternatives to Mistral Medium 3.5?›
GPT-6.1 Sol (cheaper), GPT-6 Luna (better), Qwen3.8 Flash (open weights).