Processing or generating 100,000 words (approximately 133,300 tokens) with DeepSeek V3 (Chat) ranges from $0.001866 (cached input) to $0.0373 (full generation).
You send 100,000 words as prompt context, documentation, or background knowledge.
DeepSeek V3 (Chat) 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 |
|---|---|---|---|---|---|
| DeepSeek V3 (Chat) (Current) | deepseek | $0.0187 | $0.001866 | $0.0373 | 64,000 |
| GPT-5.6 Luna | openai | $0.0256 | $0.00256 | $0.1536 | 1,050,000 |
| GPT-5.4 mini | openai | $0.096 | $0.0096 | $0.576 | 256,000 |
| GPT-5.4 nano | openai | $0.0256 | $0.00256 | $0.16 | 128,000 |
| GPT-4o mini | openai | $0.0192 | $0.0096 | $0.0768 | 128,000 |
| Claude Haiku 4.5 | anthropic | $0.132 | $0.0132 | $0.66 | 1,000,000 |
For DeepSeek V3 (Chat) (deepseek_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 $0.0187 (or $0.001866 with prompt caching). Generating 100,000 words as output costs $0.0373. 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 $0.001866). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.009331 for input, $0.0187 for output).
At an average human reading speed of 250 words per minute, 100,000 words takes approximately 400 minutes to read. In contrast, DeepSeek V3 (Chat) can process or generate this text in seconds.