Processing or generating 2,500 words (approximately 3,333 tokens) with DeepInfra — Qwen 2.5 Coder 32B ranges from $0.000067 (cached input) to $0.0008 (full generation).
Quick answer
For 2,500 words, DeepInfra — Qwen 2.5 Coder 32B processes approximately 3,333 tokens. That costs $0.000267 as uncached input, $0.000067 with the modeled cache discount, or $0.0008 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 2,500 words as prompt context, documentation, or background knowledge.
DeepInfra — Qwen 2.5 Coder 32B drafts a complete 2,500 words article, chapter, or code module.
2,000 words input prompt + 500 words output reply.
| Model | Provider | Input Cost (2,500 words) | Cached Input Cost | Output Cost (2,500 words) | Context Limit |
|---|---|---|---|---|---|
| DeepInfra — Qwen 2.5 Coder 32B (Current) | deepinfra | $0.000267 | $0.000067 | $0.0008 | 128,000 |
| GPT-5.3 Codex | openai | $0.0056 | $0.00056 | $0.0448 | 256,000 |
| DeepSeek Coder V2.5 | deepseek | $0.000467 | $0.000047 | $0.000933 | 128,000 |
| Codestral 2501 | mistral | $0.001 | $0.00010 | $0.003 | 256,000 |
| Grok Build 0.1 | xai | $0.003333 | $0.000667 | $0.006666 | 256,000 |
| Qwen 2.5 Coder 32B | qwen | $0.000667 | $0.000067 | $0.002 | 128,000 |
For DeepInfra — Qwen 2.5 Coder 32B (qwen_bpe tokenizer), 2,500 words is approximately 3,333 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 2,500 words as input costs $0.000267 (or $0.000067 with prompt caching). Generating 2,500 words as output costs $0.0008. 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 $0.000067). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.000133 for input, $0.0004 for output).
At an average human reading speed of 250 words per minute, 2,500 words takes approximately 10 minutes to read. In contrast, DeepInfra — Qwen 2.5 Coder 32B can process or generate this text in seconds.