Processing or generating 500 words (approximately 640 tokens) with GPT-5.4 Workhorse ranges from $0.00016 (cached input) to $0.0096 (full generation).
You send 500 words as prompt context, documentation, or background knowledge.
GPT-5.4 Workhorse drafts a complete 500 words article, chapter, or code module.
400 words input prompt + 100 words output reply.
| Model | Provider | Input Cost (500 words) | Cached Input Cost | Output Cost (500 words) | Context Limit |
|---|---|---|---|---|---|
| GPT-5.4 Workhorse (Current) | openai | $0.0016 | $0.00016 | $0.0096 | 256,000 |
| GPT-5.6 Terra | openai | $0.00128 | $0.000128 | $0.00768 | 1,050,000 |
| o3-mini | openai | $0.000704 | $0.000352 | $0.002816 | 200,000 |
| o4-mini | openai | $0.000704 | $0.000176 | $0.002816 | 256,000 |
| o1-mini | openai | $0.000704 | $0.000352 | $0.002816 | 128,000 |
| Claude Sonnet 5 | anthropic | $0.00132 | $0.000132 | $0.0066 | 1,000,000 |
For GPT-5.4 Workhorse (o200k_base tokenizer), 500 words is approximately 640 tokens (an average ratio of 1.28 tokens per word in English). Code, technical vocabulary, and non-English scripts will have higher token densities.
Sending 500 words as input costs $0.0016 (or $0.00016 with prompt caching). Generating 500 words as output costs $0.0096. 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 $0.00016). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.0008 for input, $0.0048 for output).
At an average human reading speed of 250 words per minute, 500 words takes approximately 2 minutes to read. In contrast, GPT-5.4 Workhorse can process or generate this text in seconds.