Processing or generating 100,000 words (approximately 133,300 tokens) with Codestral 2501 ranges from $0.003999 (cached input) to $0.12 (full generation).
You send 100,000 words as prompt context, documentation, or background knowledge.
Codestral 2501 drafts a complete 100,000 words article, chapter, or code module.
80,000 words input prompt + 20,000 words output reply.
| Model | Provider | Input Cost (100,000 words) | Cached Input Cost | Output Cost (100,000 words) | Context Limit |
|---|---|---|---|---|---|
| Codestral 2501 (Current) | mistral | $0.04 | $0.003999 | $0.12 | 256,000 |
| GPT-5.3 Codex | openai | $0.224 | $0.0224 | $1.792 | 256,000 |
| DeepSeek Coder V2.5 | deepseek | $0.0187 | $0.001866 | $0.0373 | 128,000 |
| Grok Build 0.1 | xai | $0.1333 | $0.0267 | $0.2666 | 256,000 |
| Qwen 2.5 Coder 32B | qwen | $0.0267 | $0.002666 | $0.08 | 128,000 |
| DeepInfra — Qwen 2.5 Coder 32B | deepinfra | $0.0107 | $0.002666 | $0.032 | 128,000 |
For Codestral 2501 (mistral_bpe tokenizer), 100,000 words is approximately 133,300 tokens (an average ratio of 1.33 tokens per word in English). Code, technical vocabulary, and non-English scripts will have higher token densities.
Sending 100,000 words as input costs $0.04 (or $0.003999 with prompt caching). Generating 100,000 words as output costs $0.12. Output tokens are more expensive because autoregressive token generation requires significantly more computation per token.
Yes! If your input text is part of a repeated context, prompt caching saves 50% to 90% (costing $0.003999). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.02 for input, $0.06 for output).
At an average human reading speed of 250 words per minute, 100,000 words takes approximately 400 minutes to read. In contrast, Codestral 2501 can process or generate this text in seconds.