Model comparison
Magistral Small vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 30.2 on the Noometry Index.
Last verified . 0 shared benchmarks.
Mistral AI
PerplexitySummary
- The widest gap is in reasoning, where Sonar leads 21.1 to 6.8.
- Magistral Small is cheaper at $0.50 / $1.50 per million input/output tokens, against $1 / $1 for Sonar.
- Magistral Small has downloadable open weights; the other is API-only.
Side by side
| Magistral Small | Sonar | |
|---|---|---|
| Provider | Mistral AI | Perplexity |
| Noometry Index | 30.2 | 38.5 |
| Released | 2025-06-10 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 128K | 128K |
| Max output | 40K | 4K |
| Input $ / M tokens | $0.50 | $1 |
| Output $ / M tokens | $1.50 | $1 |
| Results tracked | 10 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Small leads
Magistral Small: 38.4 (#176), Sonar: 35.7 (#221)
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| SciCode | 35.2% | — |
| LiveBench Coding | — | 35.1% |
Reasoning Sonar leads
Magistral Small: 6.8 (#350), Sonar: 21.1 (#227)
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 6.3% | — |
| ARC-AGI-1 | 5% | — |
| CritPt | 0.3% | — |
| Chess Puzzles | 3% | — |
| LiveBench Reasoning | — | 46.3% |
| DTBench | 61.3% | — |
| LiveBench Data Analysis | — | 37.9% |
| Epoch Capabilities Index | 133.19 | — |
| LiveBench | — | 46.9% |
Math Sonar leads
Magistral Small: 26.2 (#261), Sonar: 33.7 (#200)
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 30% | — |
| LiveBench Math | — | 41.6% |
Knowledge Not comparable
Magistral Small: 30.9 (#223), Sonar: —
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| GPQA Diamond | 56.1% | — |
Instruction Following Not comparable
Magistral Small: —, Sonar: 71.4 (#150)
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
Writing & Preference Not comparable
Magistral Small: —, Sonar: 52.6 (#167)
| Benchmark | Magistral Small | Sonar |
|---|---|---|
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Magistral Small better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 30.2 on the Noometry Index.
Which is cheaper, Magistral Small or Sonar?
Magistral Small is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Sonar lists at $1 and $1.
Is Magistral Small or Sonar better for coding?
Magistral Small scores higher on coding benchmarks: 38.4 versus 35.7 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do Magistral Small and Sonar share?
0 benchmarks have published results for both models. Magistral Small has 10 scored results on Noometry and Sonar has 7.