Ministral 3 8B 2512
Ministral 3 8B 2512Alternatives?
CheaperLing 3.0 FlashthinkingOpen×5.6-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 76 GB+1
BetterGPT-6 Lunalow×4.7-59%$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking×4.3+250%$0.30 / $1.20No image inputCheaperLing 3.0 Flash FinthinkingOpen×5.3-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 76 GB+1CheaperLing 3.0 Flash VLthinkingOpen×5.2-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 75 GBLing 3.0 FlashthinkingOpen×5.6-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 76 GB+1Ling 3.0 Flash FinthinkingOpen×5.3-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 76 GB+1Ling 3.0 Flash VLthinkingOpen×5.2-40%$0.06 / $0.18Memory (4-bit): 4.5 GB → 75 GBGPT-6 Lunalow×4.7-59%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×4.7-11%$0.01GPT-6 Lunalow×4.7-59%$0.005Function calling only with none · Chat Completions+3
Granite 4.2 3BthinkingOpen+55%-46%$0.006Context: 262K → 128K+1GPT-5.6 Lunalow×4.7-11%$0.01GPT OSS 20BhighOpen×2.6+3%$0.011Context: 262K → 131K+2Mistral Small 4thinkingOpen×2.3+36%$0.015Memory (4-bit): 4.5 GB → 72 GB
Gemma 4 12BthinkingOpen×3.7same$0.10 / $0.30Context: 262K → 256K+1
Qwen 3.5 9BthinkingOpen×3.5+1,818%$0.21Memory (4-bit): 4.5 GB → 4.8 GBIK2 Horizon 3.7BthinkingOpen×3.3+250%$0.30 / $1.20No image inputSApriel v1.5 15B ThinkerOpen×2.8——Context: 262K → 128K+1Mistral Large 4thinking×8.2+10,130%$1.13Infinite whitespace stream+4Mistral Medium 3.5highOpen×3.2+4,428%$0.50Memory (4-bit): 4.5 GB → 77 GBMistral Small 4thinkingOpen×2.3+36%$0.015Memory (4-bit): 4.5 GB → 72 GBMagistral Medium 1thinking×2.3——Context: 262K → 40K+1Magistral Small 1thinkingOpen+77%—$0.70 / $1.16Context: 262K → 128K+2Ministral 3 3B 2512Open+26%-29%$0.008Context: 262K → 131KMinistral 3 14B 2512non-thinkingOpensame+75%$0.019Memory (4-bit): 4.5 GB → 7 GBMistral Large 3non-thinkingOpensame+181%$0.031Memory (4-bit): not verifiedAll Mistral models →Run it yourself
estimateBenchmarks
in Chat & KnowledgeFAQ
How much does Ministral 3 8B 2512 cost?›
$0.15 per 1M input tokens and $0.15 per 1M output tokens; cached input $0.015 per 1M. A typical benchmark task costs about $0.011 at default effort.
Is Ministral 3 8B 2512 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 Ministral 3 8B 2512, and what is it best at?›
#298 of 334 in Chat & Knowledge, #75 of 132 in Coding, #119 of 133 in Agents & Tool Use, #214 of 245 in Writing & Language, #114 of 124 in Vision & Documents.
Is Ministral 3 8B 2512 better than Qwen3 4B?›
On Humanity's Last Exam, Ministral 3 8B 2512 scores 4.3 vs 4.4 — about the same.
How can I use Ministral 3 8B 2512?›
Through the API of Mistral.
What is the context window of Ministral 3 8B 2512?›
262K tokens, with up to 210K output tokens.
When was Ministral 3 8B 2512 released? Is there a newer version?›
Released 2025-12-02. It is the current version in its line.
Is Ministral 3 8B 2512 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Ministral 3 8B 2512 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Ministral 3 8B 2512 read images?›
It accepts text and images as input.
What is Ministral 3 8B 2512?›
Ministral 3 8B 2512 is an open-weights language model from Mistral, released Dec 2025.
What are the best alternatives to Ministral 3 8B 2512?›
Ling 3.0 Flash (cheaper), GPT-6 Luna (better), K2 Horizon 7B (open weights).