Processing or generating 50,000 words (approximately 66,650 tokens) with Gemma 4 31B Instruct ranges from $0.004166 (cached input) to $0.05 (full generation).
You send 50,000 words as prompt context, documentation, or background knowledge.
Gemma 4 31B Instruct drafts a complete 50,000 words article, chapter, or code module.
40,000 words input prompt + 10,000 words output reply.
| Model | Provider | Input Cost (50,000 words) | Cached Input Cost | Output Cost (50,000 words) | Context Limit |
|---|---|---|---|---|---|
| Gemma 4 31B Instruct (Current) | $0.0167 | $0.004166 | $0.05 | 128,000 | |
| GPT-5.6 Terra | openai | $0.128 | $0.0128 | $0.768 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $0.16 | $0.016 | $0.96 | 256,000 |
| o3-mini | openai | $0.0704 | $0.0352 | $0.2816 | 200,000 |
| o4-mini | openai | $0.0704 | $0.0176 | $0.2816 | 256,000 |
| o1-mini | openai | $0.0704 | $0.0352 | $0.2816 | 128,000 |
For Gemma 4 31B Instruct (sentencepiece tokenizer), 50,000 words is approximately 66,650 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 50,000 words as input costs $0.0167 (or $0.004166 with prompt caching). Generating 50,000 words as output costs $0.05. 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 $0.004166). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.008331 for input, $0.025 for output).
At an average human reading speed of 250 words per minute, 50,000 words takes approximately 200 minutes to read. In contrast, Gemma 4 31B Instruct can process or generate this text in seconds.