Mistral AILanguage model
Magistral Small
Magistral Small has no overall index value yet because there are too few independent benchmark results.
PopularityRank 76 of 785
53out of 100same as last month
based on two data groups, mostly usage
Summary
Magistral Small was Mistral’s first small reasoning model. The 24B checkpoint combines multi-step reasoning, mathematics, coding, and domain tasks. Mistral introduced it as a transparent multilingual offering under Apache 2.0; reasoning still produces longer outputs and requires more computation.
Magistral Small 1.2 revises the same product line. Current documentation lists the first Small version as retired.
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 14 published benchmark results
Page 1 of 2
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
CritPt v2: Overall (Show measurement, test conditions, and source)
- Source value
- 0
- Score
- 0
- Metric
- pass@1 accuracy
- Unit
- %
- Benchmark version
- v2
- Category
- reasoning
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- disputed
- Retrieved at
- 2026-11-02
- Methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. The full CritPt v2 run covers 70 questions with five repeats. The official grading server reports pass@1.. Source model ID: artificial-analysis/0fec07d5-a9b2-407a-b5f8-5bf10bd86b59/magistral-small/magistral-small. Source model name: Magistral Small. Provider reported by source: Mistral. Model snapshot: Magistral Small 1. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Python functions, symbolic expressions, and numerical answers; official grading server. Attempts: 70 questions, five repeats per question, pass@1. Reasoning: reasoning enabled. Score display: 0.0%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-02-67d46be20a5b. Source payload hash: 67d46be20a5b412db6cfb60ad190441fcaa5513500b8fd47d033168d0e97b6b2. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-11-02. Source cohort: artificial-analysis-critpt-v2:v2:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. The full CritPt v2 run covers 70 questions with five repeats. The official grading server reports pass@1.. Tool policy: Python functions, symbolic expressions, and numerical answers; official grading server. Attempts: 70 questions, five repeats per question, pass@1. Reasoning: reasoning enabled
- Context
- This AA source field contains the original CritPt score. The earlier CritPt v2 attribution was incorrect; the value and former label remain as disputed historical evidence.. The source dataset marks this result as disputed. It remains visible as documented evidence, but is not reliable comparative proof.
IFBench: Overall (Show measurement, test conditions, and source)
- Source value
- 24.7619047619048
- Score
- 24.7619047619048
- Metric
- IFBench prompt-level loose accuracy (pass@1)
- Unit
- %
- Benchmark version
- AA test set: 294 questions
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- superseded
- Retrieved at
- 2026-11-02
- Methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Source model ID: artificial-analysis/0fec07d5-a9b2-407a-b5f8-5bf10bd86b59/magistral-small/magistral-small. Source model name: Magistral Small. Provider reported by source: Mistral. Model snapshot: Magistral Small 1. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Score display: 24.8%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-02-67d46be20a5b. Source payload hash: 67d46be20a5b412db6cfb60ad190441fcaa5513500b8fd47d033168d0e97b6b2. Parser version: artificial-analysis-rsc-models-v3. Imported at: 2026-11-02. Source cohort: artificial-analysis-ifbench:AA test set: 294 questions:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). AA source value for average reasoning tokens before the answer: 2000. Its methodology averages 60 varied prompts and assumes 2,000 when no average is available. This is not a token cap. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled
- Context
- The often cited 58 refers to verifiable out-of-domain constraint types, not the number of questions in AA's test run.. This result is retained as historical evidence and is not interchangeable with current results.
IFBench: Overall (Show measurement, test conditions, and source)
- Source value
- 24.7619047619048
- Score
- 24.7619047619048
- Metric
- IFBench prompt-level loose accuracy (pass@1)
- Unit
- %
- Benchmark version
- AA test set: 294 questions
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-07
- Methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Source model ID: artificial-analysis/0fec07d5-a9b2-407a-b5f8-5bf10bd86b59/magistral-small/magistral-small. Source model name: Magistral Small. Provider reported by source: Mistral. Model snapshot: Magistral Small 1. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Score display: 24.8%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-07-b96e3ffc8c5b. Source payload hash: b96e3ffc8c5b052aa56f96d05e89d02b8cc234a75085820fca0ca2a39e498274. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-07. Source cohort: artificial-analysis-ifbench:AA test set: 294 questions:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Magistral Small 1 (release magistral-small, row magistral-small, UUID 0fec07d5-a9b2-407a-b5f8-5bf10bd86b59, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled
- Context
- The often cited 58 refers to verifiable out-of-domain constraint types, not the number of questions in AA's test run.
AIME 2024 (Show measurement, test conditions, and source)
- Source value
- 70.68
- Score
- 70.68
- Metric
- pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Mistral AI Magistral Small model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Exact Magistral Small row in the publisher's benchmark table; sampling and reasoning effort are not reported.
AIME 2025 (Show measurement, test conditions, and source)
- Source value
- 62.76
- Score
- 62.76
- Metric
- pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Mistral AI Magistral Small model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Exact Magistral Small row in the publisher's benchmark table; sampling and reasoning effort are not reported.
GPQA Diamond (Show measurement, test conditions, and source)
- Source value
- 68.18
- Score
- 68.18
- Metric
- score as published
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Mistral AI Magistral Small model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Exact Magistral Small row in the publisher's benchmark table; sampling and reasoning effort are not reported.
LiveCodeBench (Show measurement, test conditions, and source)
- Source value
- 55.84
- Score
- 55.84
- Metric
- score as published
- Unit
- %
- Benchmark version
- v5
- Category
- coding
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Mistral AI Magistral Small model card
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Vendor-reported result. Exact Magistral Small row; the table labels LiveCodeBench version 5. Sampling and reasoning effort are not reported.
AIME 2024 (Show measurement, test conditions, and source)
- Source value
- 83.3
- Score
- 83.3
- Metric
- majority voting @64
- Unit
- %
- Benchmark version
- AIME 2024
- Category
- mathematics
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Mistral AI, Magistral announcement (June 10, 2025)
- Status
- active
- Retrieved at
- 2026-10-19
- Methodology
- Vendor-reported result. Magistral Small; AIME 2024; majority voting @64, as stated in the announcement. The source does not specify prompts, sampling or decoding settings, task-level sample count, tie handling, checkpoint hash, or uncertainty intervals.
- Context
- The announcement does not state the prompt, sampling or decoding settings, task-level sample count, tie-breaking rule, checkpoint hash, or confidence interval.
BRIDGE clinical text understanding, zero-shot (Show measurement, test conditions, and source)
- Source value
- 37.6
- Score
- 37.6
- Metric
- Zero-shot score
- Unit
- %
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Wu et al., Why Chain of Thought Fails in Clinical Text Understanding
- Status
- active
- Retrieved at
- 2026-10-22
- Methodology
- Independent evaluation. Exact checkpoint Magistral-Small-2506; BRIDGE Table S1 zero-shot result; deterministic decoding at temperature 0 with sampling disabled. Study/result reference: BRIDGE clinical-text study, Table S1 (arXiv:2509.21933)
- Context
- The table places bracketed values beside the scores without explicitly defining them. The study describes 1,000 bootstrap resamples but does not state the nominal coverage or a model-specific sample count.
BRIDGE clinical text understanding, CoT (Show measurement, test conditions, and source)
- Source value
- 34.2
- Score
- 34.2
- Metric
- CoT score
- Unit
- %
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Wu et al., Why Chain of Thought Fails in Clinical Text Understanding
- Status
- active
- Retrieved at
- 2026-10-22
- Methodology
- Independent evaluation. Exact checkpoint Magistral-Small-2506; BRIDGE Table S1 CoT result; deterministic decoding at temperature 0 with sampling disabled. Study/result reference: BRIDGE clinical-text study, Table S1 (arXiv:2509.21933)
- Context
- The table places bracketed values beside the scores without explicitly defining them. The study describes 1,000 bootstrap resamples but does not state the nominal coverage or a model-specific sample count.
AIME (Show measurement, test conditions, and source)
- Source value
- 76.7
- Score
- 76.7
- Metric
- maj@64
- Unit
- No unit provided
- Benchmark version
- AIME'25
- Direction
- No direction published
- Source type
- vendor-reported
- Evaluator
- Mistral AI, Magistral paper (arXiv:2506.10910v1)
- Status
- active
- Retrieved at
- 2026-10-24
- Methodology
- Vendor-reported result. Magistral paper evaluation setup: temperature 0.7, top-p 1.0 for math evaluations, maximum length 40k. Table 3 does not state per-result N, replicate count, or CI; the 64/16-run note applies to Table 2 only.
- Context
- Table 3 does not print a unit, per-result sample N, replicate count, or CI. The paper’s 64 AIME / 16 LiveCodeBench averaging note explicitly describes Table 2 (Magistral Medium), not these Table 3 Magistral Small rows.
MATH-500 (Show measurement, test conditions, and source)
- Source value
- 95.9
- Score
- 95.9
- Metric
- score as reported
- Unit
- No unit provided
- Direction
- No direction published
- Source type
- vendor-reported
- Evaluator
- Mistral AI, Magistral paper (arXiv:2506.10910v1)
- Status
- active
- Retrieved at
- 2026-10-24
- Methodology
- Vendor-reported result. Magistral paper evaluation setup: temperature 0.7, top-p 1.0 for math evaluations, maximum length 32k. Table 3 does not state per-result N, replicate count, or CI; the 64/16-run note applies to Table 2 only.
- Context
- Table 3 does not print a unit, per-result sample N, replicate count, or CI. The paper’s 64 AIME / 16 LiveCodeBench averaging note explicitly describes Table 2 (Magistral Medium), not these Table 3 Magistral Small rows.
Page 1 of 2
Popularity compared
How often the model is searched for, used and mentioned, compared with every model of its category.
Popularity index
The 100 most popular models of this category, 0 to 100, higher is more popular
20 of 100 models
As of November 2026 · Source: gradually.ai popularity index
Over time
Weekly values from October 18, 2026 to November 7, 2026
Magistral SmallNemotron 3.5 Super VL 120B A12B 53Spark-X2.5-4B 52Gemini 3.7 Flash 52DeepSeek-R1-Distill-Qwen-32B 52
Method changes in the timeline, a jump at these points is no market event:
- 1November 1, 2026: Value only with search or usage data, 30-day attention
What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.
Similar models
More models of the same kind as Magistral Small, chosen by model line, developer, size, price and release date.
- Devstral SmallMistral AI · same developer
- Voxtral Small 24BMistral AI · same developer
- Magistral Small 1.2Mistral AI · same developer
- Voxtral Small (latest)Mistral AI · same developer
- Devstral Small 2Mistral AI · same developer
- Ministral 3 14BMistral AI · same developer
- Codestral-22B-v0.1Mistral AI · same developer
- Codestral (latest)Mistral AI · same developer
All specifications
Architecture
Input and output
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
Frequently asked questions about Magistral Small
How large is the context window of Magistral Small?
131,072 tokens. That is how much text Magistral Small reads at once, about 200 A4 pages.
Page count as a rule of thumb from Artificial Analysis
GLM-5.1200K
DeepSeek-V3 0324164K
Magistral SmallRank 94 of 103131K
GLM-4.5131K
GPT OSS 120B131K
DeepSeek-V3.2128K
DeepSeek-R1128K
Can I run Magistral Small locally?
Yes, the weights are open.
Stated by Mistral AI
- Open weights
- yes
- License
- Apache 2.0
- Parameters
- 24B
- active per token
- n/a
Sources and data date
Show all 14 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
| Evaluator | Test series | Composite indices | Recorded active measurements |
|---|---|---|---|
| Artificial Analysis | 1 | 0 | 1 |
Without vendor claims and historical or disputed results; those stay labelled in the single values.