Processing or generating 1,000,000 words (approximately 1,333,000 tokens) with DeepInfra — Qwen 2.5 Coder 32B ranges from $0.0267 (cached input) to $0.3199 (full generation).
Quick answer
For 1,000,000 words, DeepInfra — Qwen 2.5 Coder 32B processes approximately 1,333,000 tokens. That costs $0.1066 as uncached input, $0.0267 with the modeled cache discount, or $0.3199 when generated as output.
Method & trust
The estimate converts the selected English word count with the model tokenizer family, then applies the published input/output price per million tokens. Real prompts vary with code, formatting, language, and system instructions.
You send 1,000,000 words as prompt context, documentation, or background knowledge.
DeepInfra — Qwen 2.5 Coder 32B drafts a complete 1,000,000 words article, chapter, or code module.
800,000 words input prompt + 200,000 words output reply.
| Model | Provider | Input Cost (1,000,000 words) | Cached Input Cost | Output Cost (1,000,000 words) | Context Limit |
|---|---|---|---|---|---|
| DeepInfra — Qwen 2.5 Coder 32B (Current) | deepinfra | $0.1066 | $0.0267 | $0.3199 | 128,000 |
| GPT-5.3 Codex | openai | $2.24 | $0.224 | $17.92 | 256,000 |
| DeepSeek Coder V2.5 | deepseek | $0.1866 | $0.0187 | $0.3732 | 128,000 |
| Codestral 2501 | mistral | $0.3999 | $0.04 | $1.20 | 256,000 |
| Grok Build 0.1 | xai | $1.333 | $0.2666 | $2.666 | 256,000 |
| Qwen 2.5 Coder 32B | qwen | $0.2666 | $0.0267 | $0.7998 | 128,000 |
For DeepInfra — Qwen 2.5 Coder 32B (qwen_bpe tokenizer), 1,000,000 words is approximately 1,333,000 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 1,000,000 words as input costs $0.1066 (or $0.0267 with prompt caching). Generating 1,000,000 words as output costs $0.3199. 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 $0.0267). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.0533 for input, $0.16 for output).
At an average human reading speed of 250 words per minute, 1,000,000 words takes approximately 4000 minutes to read. In contrast, DeepInfra — Qwen 2.5 Coder 32B can process or generate this text in seconds.