Mistral Small 4 (Reasoning) API Pricing
Mistral · released 2026-03-16 · merges 2 reasoning-effort tiers
Benchmarks shown for the highest-scoring reasoning tier. Merged tiers: Mistral Small 4 (Reasoning), Mistral Small 4 (Non-reasoning).
Pricing
0.015¢ / 1K
0.060¢ / 1K
0.026¢ / 1K
What it actually costs
| Workload | Tokens | Cost |
|---|---|---|
| Short chat reply | 500 in / 300 out | 0.0255¢ |
| RAG query | 4,000 in / 500 out | 0.09¢ |
| Long document summary | 50,000 in / 1,000 out | 0.81¢ |
| Code review of a 2k-line file | 30,000 in / 2,000 out | 0.57¢ |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $7.65 |
Cost = (input tokens ÷ 1,000,000 × $0.15) + (output tokens ÷ 1,000,000 × $0.60).
Benchmarks
#235 of 533
#119 of 197
Similar intelligence, lower price
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Gemma 4 E4B (Reasoning) | 8.9 | $0.04 |
| Granite 4.2 3B | 9.1 | $0.052 |
| Qwen3.5 4B (Reasoning) | 13.1 | $0.06 |
| Sarvam 105B (high) | 8.8 | $0.074 |
| NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | 8.9 | $0.088 |
Leaderboard neighbours
| Model | Intelligence | Blended $/1M |
|---|---|---|
| HyperCLOVA X SEED Think (32B) | 11.4 | — |
| o1-preview | 11.4 | $28.875 |
| Grok 4.1 Fast (Non-reasoning) | 11.3 | — |
| Mistral Small 4 (Reasoning) | 11.3 | $0.262 |
| GLM-4.6V (Reasoning) | 11.2 | $0.45 |
| Qwen3 Next 80B A3B (Reasoning) | 11.2 | $0.412 |
| GLM-4.5-Air | 11.1 | $0.372 |
More from Mistral
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Mistral Medium 3.5 | 14.2 | $3.00 |
| Magistral Medium 1.2 | 11.8 | — |
| Mistral Large 3 | 9.3 | $0.75 |
| Mistral Medium 3.1 | 9.2 | $0.80 |
| Magistral Medium 1 | 9.1 | — |
| Devstral Medium | 9.0 | — |
| Mistral Medium 3 | 9.0 | $0.80 |
| Devstral Small (May '25) | 8.7 | — |
Count tokens & estimate cost for Mistral Small 4 (Reasoning) See Mistral Small 4 (Reasoning) on the leaderboard
Source: Artificial Analysis · last synced 2026-09-22
FAQ
Who develops Mistral Small 4 (Reasoning)?
Mistral Small 4 (Reasoning) is developed by Mistral, released 2026-03-16.
How much does Mistral Small 4 (Reasoning) cost per million tokens?
Mistral Small 4 (Reasoning) costs $0.15 per 1M input tokens and $0.60 per 1M output tokens ($0.262 blended), per Artificial Analysis's official pricing data.
How does Mistral Small 4 (Reasoning) rank on benchmarks?
Mistral Small 4 (Reasoning) scores 11.3 on the Artificial Analysis Intelligence Index, ranking #235 of 533 models we track.
Is there a cheaper model with similar intelligence to Mistral Small 4 (Reasoning)?
Yes — Gemma 4 E4B (Reasoning) scores similar intelligence (8.9) at $0.04/1M blended, versus $0.262/1M for Mistral Small 4 (Reasoning).
Where does this data come from?
Pricing and benchmark data for Mistral Small 4 (Reasoning) comes from Artificial Analysis (artificialanalysis.ai), synced daily. Last synced 2026-09-22.