Codestral vs DeepSeek-V4-Flash
DeepSeek-V4-Flash is cheaper for every workload modelled here.
List prices checked
Specifications and rates
| Property | Codestral | DeepSeek-V4-Flash |
|---|---|---|
| Input per 1M | $0.300 | $0.140 |
| Output per 1M | $0.900 | $0.280 |
| Cached input per 1M | — | $0.0028 |
| Blended (3:1) | $0.450 | $0.175 |
| Context window | 262K | 1M |
| Max output | — | 384K |
| Inputs accepted | text | text |
Source: Mistral AI pricing (May 2026) · DeepSeek pricing (August 2026). Published list prices only — committed-use discounts, regional variation, and per-request charges are not included. Not yet independently re-checked against the vendor’s page.
Monthly cost by workload
Identical token volumes through both rate cards. Output-heavy profiles are where the ranking flips.
| Workload | Codestral | DeepSeek-V4-Flash | Cheaper |
|---|---|---|---|
| In-app chat assistantA support or help assistant handling a few thousand conversations a month, with short prompts and short answers. | $11 | $4.20 | DeepSeek-V4-Flashsaves $6.30/mo (60%) |
| RAG over documentsRetrieval-augmented answering where each request stuffs several retrieved passages into the prompt, so input dominates. | $104 | $46 | DeepSeek-V4-Flashsaves $57/mo (55%) |
| Coding agentAn agentic loop that reads files and writes patches, generating heavy output and re-sending large context on every turn. | $210 | $84 | DeepSeek-V4-Flashsaves $126/mo (60%) |
What each one is for
Codestral
by Mistral AI
A code-specialised model tuned for fill-in-the-middle completion, priced for the high request volume that inline editor autocomplete generates.
DeepSeek-V4-Flash
by DeepSeek
A million-token context window at a fraction of any Western model's price, with cache-hit input billed at a fiftieth of the base rate. DeepSeek has published notice of a significant price rise, so treat these rates as a floor rather than a plan.
Common questions
- Is Codestral or DeepSeek-V4-Flash cheaper in 2026?
- DeepSeek-V4-Flash, for every workload modelled here. On a blended 3:1 basis it costs $0.175 per million tokens against Codestral's $0.450.
- Which has the larger context window?
- DeepSeek-V4-Flash, at 1M tokens against 262K. A larger window only helps if you can afford to fill it — the cost of one full-context request is in the table above.
- Which is cheaper for output-heavy work like agents?
- DeepSeek-V4-Flash, at $0.280 per million output tokens against $0.900. Reasoning tokens bill as output, so for agentic loops this axis usually decides the invoice regardless of what the input rates look like.
- What is not included in these figures?
- Batch-processing discounts, fine-tuning, long-context surcharges above a provider's standard tier, and any enterprise agreement. These are published pay-as-you-go list prices, checked against each provider's own pricing page on the date shown.
- Is switching from Codestral to DeepSeek-V4-Flash worth it?
- On price alone it is about 61% cheaper on a blended basis. Whether that survives contact with your workload depends on output quality and on retry rate — a cheaper model that needs two attempts is not cheaper. Model the specific volumes in the calculator before you migrate.