vs
0:3
overall capability wins (0 ties across 3 scored). Mistral Small 4 leads overall.
- Reasoning0:1
- Pricing0:3
- Speed0:2
- Specs0:2
Verdict
Mistral Small 4 edges reasoning & knowledge; Mistral Small 4 offers ~2.0x larger context (256K vs 128K); Mistral Small 4 allows ~1.6x more max output (64K vs 40K).
- Mistral Small 4 edges reasoning & knowledge
- Mistral Small 4 offers ~2.0x larger context (256K vs 128K)
- Mistral Small 4 allows ~1.6x more max output (64K vs 40K)
- Mistral Small 4 is ~2.9x cheaper on a blended token basis than Magistral Small
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Magistral Small
- $0.88
- Mistral Small 4
- $0.30
Mistral Small 4 is about 2.9x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Mistral
Mistral
Family
Mistral
Mistral
License
Apache 2.0
Apache 2.0
Open weights
Yes
Yes
Release
Jun 10, 2025
Mar 1, 2026
Knowledge cutoff
-
-
API / provider
Mistral
Mistral
Modalities
text, image → text
text, image → text
Specs
Context window
128K
256K
Max output
40K
64K
Parameters
24B
—
Pricing
Input $/1M
$0.50
$0.15
Output $/1M
$1.50
$0.60
Blended $/1M (3∶1)
$0.75
$0.26
Speed
tok/s
90
140
TTFT (s)
0.55
0.2
Reasoning
MMLU-Pro
—
—
GPQA Diamond
68.2
76.9
Humanity's Last Exam
—
9.9
AIME 2025
62.8
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
21.7
AA-LCR v1.1
—
49.7
CritPt
—
0.3
MMMU-Pro
—
56.8
IFBench
—
48.2
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
55.8
—
Terminal-Bench 2.1
—
21.0
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
—
38.8
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
—
—
| Metric | Magistral Small | Mistral Small 4 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Mistral | Mistral | — |
| Family | Mistral | Mistral | — |
| License | Apache 2.0 | Apache 2.0 | — |
| Open weights | Yes | Yes | — |
| Release | Jun 10, 2025 | Mar 1, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Mistral | Mistral | — |
| Modalities | text, image → text | text, image → text | — |
| Specs | |||
| Context window | 128K | 256K | B +128K |
| Max output | 40K | 64K | B +24K |
| Parameters | 24B | — | — |
| Pricing | |||
| Input $/1M | $0.50 | $0.15 | B +$0.35 |
| Output $/1M | $1.50 | $0.60 | B +$0.90 |
| Blended $/1M (3∶1) | $0.75 | $0.26 | B +$0.49 |
| Speed | |||
| tok/s | 90 | 140 | B +50 |
| TTFT (s) | 0.55 | 0.2 | B +0.35 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 68.2 | 76.9 | B +8.7 pts |
| Humanity's Last Exam | — | 9.9 | — |
| AIME 2025 | 62.8 | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | 21.7 | — |
| AA-LCR v1.1 | — | 49.7 | — |
| CritPt | — | 0.3 | — |
| MMMU-Pro | — | 56.8 | — |
| IFBench | — | 48.2 | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 55.8 | — | — |
| Terminal-Bench 2.1 | — | 21.0 | — |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | — | 38.8 | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | — | — | — |
Magistral Small
- Context
- 128K / 40K out
- Parameters
- 24B
- Price
- $0.50 / $1.50
- Speed
- 90 tok/s · 0.55s TTFT
- Modalities
- text, image → text
- License
- Apache 2.0
Open-weight 24B Magistral reasoning model (Apache 2.0); deprecated on Mistral API in favor of Small 4 / Medium 3.5 reasoning modes.
Mistral Small 4
- Context
- 256K / 64K out
- Parameters
- —
- Price
- $0.15 / $0.60
- Speed
- 140 tok/s · 0.2s TTFT
- Modalities
- text, image → text
- License
- Apache 2.0
Hybrid instruct/reasoning/coding Small model under Apache 2.0 - Mistral's high-throughput production default.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.