// AI Platform Drift
Mistral Large 4 context length doubled to 1,048,576 tokens
OpenRouter model catalogue · published 2026-10-08 23:23 UTC
WHAT CHANGEDOpenRouter's catalogue now shows Mistral Large 4 with a context_length of 1,048,576 tokens, up from 524,288 tokens. Pricing remains $0.68 per million prompt tokens and $2.09 per million completion tokens. Max completion tokens stays at 262,144.
WHY IT MATTERSPrompts that previously hit the 524,288-token limit can now process twice as much input in a single request. Existing code that chunks documents or conversations to stay under the old limit may be doing unnecessary work.
WHAT TO DOGrep your codebase for hardcoded limits on Mistral Large 4 input size—look for 524288, 500000, or similar values in chunking logic. If you're splitting documents or context windows to fit the old limit, you can now send larger inputs in one call.
Model-reported confidence 95% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
name: Mistral: Mistral Large 4 pricing.prompt: 0.00000068 pricing.completion: 0.00000209 -context_length: 524288 +context_length: 1048576 top_provider.max_completion_tokens: 262144
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.95, needed 0.5 |
|---|---|
| triage reasoning | Context window expansion materially changes what workloads the model can handle. |
| signals | context_length doubled: 524288 → 1048576 tokens · model capability change affecting max input size |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |