# DeepSeek vs Mistral: API prices compared, dated and sourced

Canonical page: https://bertrande.com/en/vs/deepseek-vs-mistral/
License: CC-BY-4.0, reuse with attribution and a link.

No value below was produced, estimated or completed by a language model. Every figure was read from a source we name, on the date we give.

The cheapest DeepSeek model we track, deepseek-flash, costs $1.20 per million output tokens and $0.300 per million input tokens. The cheapest Mistral model, Classifier API model (8B), costs $0.040 per million output tokens and $0.040 per million input tokens. At the entry level, Mistral is 30.0 times cheaper than DeepSeek on output tokens. Prices read on 11 September 2026 and 18 September 2026.

## DeepSeek models

| Model | Input $/1M | Output $/1M | Read on |
|---|---|---|---|
| deepseek-flash | $0.300 | $1.20 | 2026-09-11 |
| deepseek-v4-pro | $1.32 | $3.96 | 2026-09-11 |

## Mistral models

| Model | Input $/1M | Output $/1M | Read on |
|---|---|---|---|
| Classifier API model (8B) | $0.040 | $0.040 | 2026-09-18 |
| Classifier API model (3B) | $0.100 | $0.100 | 2026-09-18 |
| Ministral 3 (3B) | $0.100 | $0.100 | 2026-09-18 |
| Ministral 3 (8B) | $0.150 | $0.150 | 2026-09-18 |
| Ministral 3 (14B) | $0.200 | $0.200 | 2026-09-18 |
| Voxtral Small | by content type | $0.400 | 2026-09-18 |
| Mistral Small 4 | $0.150 | $0.600 | 2026-09-18 |
| Codestral | $0.300 | $0.900 | 2026-09-18 |
| Mistral Large 3 | $0.500 | $1.50 | 2026-09-18 |
| GLM 5.2 (hosted by Mistral) | $1.40 | $4.40 | 2026-09-18 |
| GLM 5.3 (hosted by Mistral) | $1.40 | $4.40 | 2026-09-17 |
| Mistral Medium 3.5 | $1.50 | $7.50 | 2026-09-18 |


Method: https://bertrande.com/en/methodology/
Report an error: https://bertrande.com/en/corrections/