Ministral 3 14B 2512
Ministral 3 14B 2512Rankingin non-thinking mode?
Alternatives?
CheaperAgnes 2.5 Pro Beta×5.8-25%$0.10 / $0.30
BetterGPT-6 Lunalow×3.6-77%$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×5.6+5,084%$1.01Memory (4-bit): 7 GB → 17 GB+5CheaperGPT-5.6 Lunamedium×4.1-20%$0.016
CheaperLing 3.0 Flash VLthinkingOpen×4.1-55%$0.06 / $0.18Memory (4-bit): 7 GB → 75 GBAgnes 2.5 Pro Beta×5.8-25%$0.10 / $0.30GPT-6 Lunamedium×5-10%$0.017Function calling only with none · Chat Completions+3GPT-5.6 Lunamedium×4.1-20%$0.016Ling 3.0 Flash VLthinkingOpen×4.1-55%$0.06 / $0.18Memory (4-bit): 7 GB → 75 GB
MiMo v2 Omnithinking×4-13%$0.14 / $0.28GPT-6 Lunalow×3.6-77%$0.005Function calling only with none · Chat Completions+3
Granite 4.2 3BthinkingOpen+50%-69%$0.006Context: 262K → 128K+1
Gemma 4 E4BthinkingOpen+47%-58%$0.008Context: 262K → 128KGPT-5.6 Lunalow×3.5-49%$0.01GPT OSS 20BhighOpen+48%-41%$0.011Context: 262K → 131K+2Qwen3.8 27BxhighOpen×5.6+5,084%$1.01Memory (4-bit): 7 GB → 17 GB+5Qwen 3.5 27BthinkingOpen×3.8+312%$0.30 / $2.40Memory (4-bit): 7 GB → 17 GBQwen3.6 27BthinkingOpen×3.5+3,096%$0.62Memory (4-bit): 7 GB → 17 GBIK2 Horizon 7BthinkingOpen×3.4+162%$0.30 / $1.20No image input
Muse Glimmer 30BhighOpen×2.9+196%$0.058Context: 262K → 131K+1Mistral Large 4thinking×6.3+5,735%$1.13Infinite whitespace stream+4Mistral Medium 3.5highOpen×2.3+2,483%$0.50Memory (4-bit): 7 GB → 77 GBMistral Small 4thinkingOpen+86%-23%$0.015Memory (4-bit): 7 GB → 72 GBMistral Large 3non-thinkingOpen+53%+60%$0.031Memory (4-bit): not verifiedMagistral Medium 1thinking+50%——Context: 262K → 40K+1Devstral Mediumnon-thinkingOpen+49%—$0.40 / $2Context: 262K → 128K+2Magistral Small 1thinkingOpen+36%—$0.70 / $1.16Context: 262K → 128K+2Mistral Small 3non-thinkingOpensame—$0.05 / $0.08Context: 262K → 32K+2All Mistral models →Run it yourself
estimateBenchmarks
all categoriesFAQ
How much does Ministral 3 14B 2512 cost?›
$0.20 per 1M input tokens and $0.20 per 1M output tokens; cached input $0.02 per 1M. A typical benchmark task costs about $0.019 at default effort.
Is Ministral 3 14B 2512 free to use?›
There is no free API; the cheapest paid price is $0.20 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Ministral 3 14B 2512, and what is it best at?›
#288 of 334 in Chat & Knowledge, #75 of 132 in Coding, #118 of 133 in Agents & Tool Use, #86 of 95 in Writing & Language, #111 of 124 in Vision & Documents.
Is Ministral 3 14B 2512 better than Granite 4.0 H Small?›
On AA Intelligence Index 4.3.2, Ministral 3 14B 2512 scores 6 vs 6 — about the same.
How can I use Ministral 3 14B 2512?›
Through the API of Mistral.
What is the context window of Ministral 3 14B 2512?›
262K tokens, with up to 210K output tokens.
When was Ministral 3 14B 2512 released? Is there a newer version?›
Released 2025-12-02. It is the current version in its line.
Is Ministral 3 14B 2512 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Ministral 3 14B 2512 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Ministral 3 14B 2512 read images?›
It accepts text and images as input.
What is Ministral 3 14B 2512?›
Ministral 3 14B 2512 is an open-weights language model from Mistral, released Dec 2025.
What are the best alternatives to Ministral 3 14B 2512?›
Agnes 2.5 Pro Beta (cheaper), GPT-6 Luna (better), Qwen3.8 27B (open weights).