Head-to-head comparison
Codestral 2508 (Mistral AI) and Qwen2.5 72B Instruct (Qwen) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.
Snapshot as of 3 September 2026
As of 3 September 2026, Qwen2.5 72B Instruct holds the stronger ModelCap Index position (#180 vs #218); Qwen2.5 72B Instruct is 2.3× cheaper per output token ($0.40/1M vs $0.90/1M); and Codestral 2508 offers the longer context window (256K vs 33K tokens).
Evidence: publisher corpus prior · leave-one-anchor-out calibrated · 3 public benchmark observations across 3 boards. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.
Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.
Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.
Qwen2.5 72B Instruct ranks higher on the ModelCap Index as of 3 September 2026: #180 against #218. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it.
Qwen2.5 72B Instruct is cheaper on output tokens: $0.40/1M against $0.90/1M. Input tokens are $0.30/1M for Codestral 2508 and $0.36/1M for Qwen2.5 72B Instruct. Prices are the lowest listed API offer ModelCap observed, in USD per million tokens.
Codestral 2508 has the larger context window: 256K tokens against 33K.
The two models do not share a coding benchmark board on ModelCap yet, so no head-to-head coding score is published; the ModelCap Index position is the closest overall signal.
Codestral 2508: API only. Qwen2.5 72B Instruct: Restricted license. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.
ModelCap currently lists 1 API provider for Codestral 2508 and 2 for Qwen2.5 72B Instruct, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.
Every figure comes from the sealed ModelCap dataset published 3 September 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.