Mistral AI, open weights
Magistral Medium
Magistral Medium by Mistral AI ranks 227th of 354 ranked models on the Noometry Index as of October 2026, with a score of 35.2. Its strongest category is coding, where it ranks 161st. API pricing starts at $2 per million input tokens and $5 per million output tokens, with a 262K-token context window.
Last verified
Specifications
- Noometry rank
- #227 of 354
- Index score
- 35.2
- Evidence
- Confirmed 22 results
- Provider
Mistral AI- Released
- March 17, 2025
- Weights
- Open weights
- Reasoning
- Yes
- Context window
- 262K
- Max output
- 16K
- Input price
- $2 / M
- Output price
- $5 / M
- Blended price
- $2.75 / M
- Output speed
- 0 tokens/s Kagi
- Value
- #173 of 219
- Knowledge cutoff
- June 2025
- Input
- text, image
Category scores
Each category score combines every public result we have in that category.
- Coding 39.1
- Reasoning 8.6
- Math 35.1
- Knowledge 33.5
- Multilingual 39.6
- Instruction Following 66.0
- Long Context 39.3
- Writing & Preference 46.3
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 39.1 | #161 | 2 |
| Reasoning | 8.6 | #348 | 5 |
| Math | 35.1 | #189 | 1 |
| Knowledge | 33.5 | #202 | 1 |
| Multilingual | 39.6 | #224 | 1 |
| Instruction Following | 66.0 | #211 | 1 |
| Long Context | 39.3 | #183 | 1 |
| Writing & Preference | 46.3 | #219 | 3 |
Strengths and weaknesses
Categories where Magistral Medium places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Coding | 39.1 | +0.4 | #161 of 340, top 48% |
| Math | 35.1 | −1.5 | #189 of 327, top 58% |
| Long Context | 39.3 | −1.6 | #183 of 296, top 62% |
Weakest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Reasoning | 8.6 | −15.0 | #348 of 350, top 100% |
| Multilingual | 39.6 | −7.8 | #224 of 297, top 76% |
| Writing & Preference | 46.3 | −7.5 | #219 of 312, top 71% |
Closest competitors
The models ranked just above and below Magistral Medium. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Nemotron 4 340b Instruct | #223 | 35.9 | — | — | Compare |
| Llama 3.1 Tulu 3 8b | #224 | 35.7 | — | — | Compare |
| Qwen3 14B | #225 | 35.5 | $0.61 | 79 | Compare |
| DeepSeek-R1-Distill-Qwen-32B | #226 | 35.5 | — | — | Compare |
| Gemini 2.0 Flash (Feb 2025) | #228 | 35.1 | — | 92 | Compare |
| C4ai Aya Expanse 8b | #229 | 34.9 | — | — | Compare |
| Qwen Max | #230 | 34.7 | $2.80 | — | Compare |
| Claude 3.5 Sonnet | #231 | 34.6 | — | — | 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.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| SciCode | 39.2% | #84 of 121, top 70% | Epoch AI | ||
| LMArena Coding | 1319 | #185 of 294, top 63% | LMArena | 2026-10-08 |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| ARC-AGI-2 | 0% | #82 of 83, top 99% | Epoch AI | ||
| Kagi LLM Benchmark | 16.2% | #98 of 99, top 99% | Kagi LLM Benchmark | ||
| ARC-AGI-1 | 6.1% | #75 of 83, top 91% | Epoch AI | ||
| CritPt | 0.3% | #94 of 134, top 71% | Epoch AI | ||
| LMArena Hard Prompts | 1267 | #209 of 297, top 71% | LMArena | 2026-10-08 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Math | 1250 | #214 of 285, top 76% | LMArena | 2026-10-08 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Expert | 1223 | #210 of 273, top 77% | LMArena | 2026-10-08 |
Multilingual
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Non-English | 1232 | #224 of 297, top 76% | LMArena | 2026-10-08 | |
| LMArena Chinese | 1227 | #219 of 285, top 77% | LMArena | 2026-10-08 | |
| LMArena French | 1267 | #177 of 223, top 80% | LMArena | 2026-10-08 | |
| LMArena German | 1248 | #179 of 231, top 78% | LMArena | 2026-10-08 | |
| LMArena Japanese | 1175 | #167 of 211, top 80% | LMArena | 2026-10-08 | |
| LMArena Korean | 1125 | #187 of 213, top 88% | LMArena | 2026-10-08 | |
| LMArena Russian | 1224 | #227 of 283, top 81% | LMArena | 2026-10-08 | |
| LMArena Spanish | 1271 | #172 of 226, top 77% | LMArena | 2026-10-08 |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Instruction Following | 1254 | #208 of 298, top 70% | LMArena | 2026-10-08 |
Long Context
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Longer Query | 1295 | #194 of 291, top 67% | LMArena | 2026-10-08 |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Text | 1255 | #226 of 297, top 77% | LMArena | 2026-10-08 | |
| LMArena Creative Writing | 1245 | #209 of 295, top 71% | LMArena | 2026-10-08 | |
| LMArena Multi-Turn | 1275 | #208 of 295, top 71% | LMArena | 2026-10-08 |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| mistral | $2 | $5 | — | 2026-10-10 |
Compare Magistral Medium
- Magistral Medium vs DeepSeek-R1-Distill-Qwen-32B
- Magistral Medium vs Gemini 2.0 Flash (Feb 2025)
- Magistral Medium vs Qwen3 14B
- Magistral Medium vs C4ai Aya Expanse 8b
- Magistral Medium vs Llama 3.1 Tulu 3 8b
- Magistral Medium vs Qwen Max
- Magistral Medium vs GPT-6 Astra
- Magistral Medium vs Claude Fable 5.1
- Magistral Medium vs Gemini 3.8 Flash
- Magistral Medium vs Kimi K3
- Magistral Medium vs Grok 4.6
- Magistral Medium vs Qwen3.8 Max
- Magistral Medium vs GLM-5.3
- Magistral Medium 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
- Devstral Small 250534.3
- Mistral Small33.4
- Pixtral Large32.2
- Mistral Large31.9
Frequently asked questions
How good is Magistral Medium?
Magistral Medium by Mistral AI ranks 227th of 354 ranked models on the Noometry Index as of October 2026, with a score of 35.2. Its strongest category is coding, where it ranks 161st. API pricing starts at $2 per million input tokens and $5 per million output tokens, with a 262K-token context window.
How much does Magistral Medium cost?
Magistral Medium costs $2 per million input tokens and $5 per million output tokens on Mistral AI's own API.
What is Magistral Medium's context window?
Magistral Medium accepts up to 262K tokens of input and can write up to 16K tokens in one response.
Is Magistral Medium open source?
Yes. Magistral Medium's weights are downloadable; check the license for commercial terms.
How fast is Magistral Medium?
Magistral Medium generated about 0 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.
What are Magistral Medium's strengths and weaknesses?
Relative to other ranked models, Magistral Medium places best in coding, math, long context and lowest in reasoning, multilingual, writing & preference.
What is Magistral Medium best at?
Its best category is coding, where it ranks 161st on Noometry.