Mistral AI, open weights
Ministral 3B
Ministral 3B by Mistral AI ranks 338th of 354 ranked models on the Noometry Index as of October 2026, with a score of 26.2. Its strongest category is math, where it ranks 258th. API pricing starts at $0.10 per million input tokens and $0.10 per million output tokens, with a 131K-token context window.
Last verified
Specifications
- Noometry rank
- #338 of 354
- Index score
- 26.2
- Evidence
- Confirmed 6 results
- Provider
Mistral AI- Released
- October 1, 2024
- Weights
- Open weights
- Reasoning
- No
- Context window
- 131K
- Max output
- 262K
- Input price
- $0.10 / M
- Output price
- $0.10 / M
- Blended price
- $0.10 / M
- Output speed
- Not measured
- Value
- #27 of 219
- Knowledge cutoff
- October 2024
- Input
- text, image
- Hugging Face
- mistralai/Ministral-3-3B-Instruct-2512
Category scores
Each category score combines every public result we have in that category.
- Reasoning 18.4
- Math 26.6
- Knowledge 10.4
Strengths and weaknesses
Categories where Ministral 3B places highest and lowest among the models ranked in each, with its score against that category's median.
Closest competitors
The models ranked just above and below Ministral 3B. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Mixtral 8x7B | #334 | 27.1 | $0.70 | — | Compare |
| Qwen Turbo | #335 | 27.1 | $0.0875 | — | Compare |
| Qwen3-1.7B | #336 | 26.6 | — | — | Compare |
| Mistral Nemo | #337 | 26.4 | $0.15 | — | Compare |
| DeepSeek-R1-Distill-Qwen-1.5B | #339 | 26.1 | — | — | Compare |
| Claude 3 Haiku | #340 | 25.9 | — | 41 | Compare |
| Gemma 2 9B | #341 | 25.9 | — | — | Compare |
| Dolly 2.0-12b | #342 | 25.5 | — | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| DTBench | 51.7% | #132 of 151, top 88% | Epoch AI | ||
| LMCA | 5.5% | #120 of 125, top 96% | Epoch AI | ||
| Epoch Capabilities Index | 118.1 | #178 of 213, top 84% | Epoch AI | 2024-10-16 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| MATH Level 5 | 14.4% | #69 of 79, top 88% | Epoch AI | 2025-01-27 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | 25.3% | #180 of 186, top 97% | Epoch AI | 2025-01-27 | |
| Vectara Hallucination Rate (lower is better) | 7.3% | #30 of 96, top 32% | Vectara Hallucination Leaderboard | ||
| Vectara Hallucination Rate (lower is better) | 24.2% | Vectara Hallucination Leaderboard |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| azure | $0.04 | $0.04 | — | 2026-10-10 |
| bedrock | $0.10 | $0.10 | — | 2026-10-10 |
| mistral | $0.10 | $0.10 | $0.01 | 2026-10-10 |
| openrouter | $0.10 | $0.10 | $0.01 | 2026-10-10 |
Compare Ministral 3B
- Ministral 3B vs Mistral Nemo
- Ministral 3B vs DeepSeek-R1-Distill-Qwen-1.5B
- Ministral 3B vs Qwen3-1.7B
- Ministral 3B vs Claude 3 Haiku
- Ministral 3B vs Qwen Turbo
- Ministral 3B vs Gemma 2 9B
- Ministral 3B vs GPT-6 Astra
- Ministral 3B vs Claude Fable 5.1
- Ministral 3B vs Gemini 3.8 Flash
- Ministral 3B vs Kimi K3
- Ministral 3B vs Grok 4.6
- Ministral 3B vs Qwen3.8 Max
- Ministral 3B vs GLM-5.3
- Ministral 3B vs Muse Spark 1.3
Other Mistral AI models
- Mistral Large 443.1
- Mistral Medium 3.540.2
- Mistral Large 339.1
- Mistral Medium36.3
- Magistral Medium35.2
- Devstral Small 250534.3
- Mistral Small33.4
- Pixtral Large32.2
Frequently asked questions
How good is Ministral 3B?
Ministral 3B by Mistral AI ranks 338th of 354 ranked models on the Noometry Index as of October 2026, with a score of 26.2. Its strongest category is math, where it ranks 258th. API pricing starts at $0.10 per million input tokens and $0.10 per million output tokens, with a 131K-token context window.
How much does Ministral 3B cost?
Ministral 3B costs $0.10 per million input tokens and $0.10 per million output tokens on Mistral AI's own API, with cached input at $0.01.
What is Ministral 3B's context window?
Ministral 3B accepts up to 131K tokens of input and can write up to 262K tokens in one response.
Is Ministral 3B open source?
Yes. Ministral 3B's weights are downloadable from Hugging Face (mistralai/Ministral-3-3B-Instruct-2512); check the license for commercial terms.
What are Ministral 3B's strengths and weaknesses?
Relative to other ranked models, Ministral 3B places best in math and lowest in knowledge.
What is Ministral 3B best at?
Its best category is math, where it ranks 258th on Noometry.