Skip to content

Qwen3-Coder-Plus vs Codestral (25.08)

Facts only. This page does not pick a winner. For a picked winner, see the use case comparison on Tools.

Qwen3-Coder-PlusCodestral (25.08)
API model IDsqwen3-coder-pluscodestral-2508
Context window1,000,000256,000
Lifecycleactiveactive
Licenseproprietarysource-available
Trains on API dataNoYes
Zero data retentionnoon-request
Input per 1M tokens$1 (verified 2026-09-08)$0.3 (verified 2026-09-08)

Frequently asked questions

Which has the larger context window, Qwen3-Coder-Plus or Codestral (25.08)?
Qwen3-Coder-Plus holds 1,000,000 tokens and Codestral (25.08) holds 256,000. Qwen3-Coder-Plus holds more. The context window counts the prompt, any attached files, and the reply together.
Which is cheaper, Qwen3-Coder-Plus or Codestral (25.08)?
Per million input tokens, Qwen3-Coder-Plus was $1 (verified 2026-09-08) and Codestral (25.08) was $0.3 (verified 2026-09-08). Codestral (25.08) is cheaper per input token. Output tokens cost $5 and $0.9 respectively, and output volume usually decides the bill.
Does Qwen3-Coder-Plus or Codestral (25.08) train on API data?
Qwen3-Coder-Plus does not train on API data by default, and Codestral (25.08) trains on API data by default. Zero data retention is no for Qwen3-Coder-Plus and on-request for Codestral (25.08). Verified 2026-09-08 and 2026-09-08 against the providers' own documentation.
Is Qwen3-Coder-Plus or Codestral (25.08) being retired?
Qwen3-Coder-Plus is active. Codestral (25.08) is active. Lifecycle checked 2026-09-08 and 2026-09-08. Pin a dated snapshot ID rather than an alias if a retirement would break you.