AI Cost Calculator / Model pricing
Mixtral 8x7b costs $0.070 per 1M input tokens and $0.280 per 1M output tokens (list pricing as of 2026-08-15, sourced from Perplexity). See the full rate card below and project your monthly spend in the calculator.
| Input | $0.070 / 1M tokens |
| Output | $0.280 / 1M tokens |
| Cache read | Not published |
| Cache write | Not published |
| Context window | 4,096 tokens |
| Max output | 4,096 tokens |
Official source: Mistral pricing page . Batch API discounts (typically 50%) are available via the calculator's Batch mode.
A chat app making 500 calls/day at 1.2K input + 400 output tokens spends roughly $2.94/mo on Mixtral 8x7b. Enter your real workload for an exact projection and to compare against other models.
Calculate Mixtral 8x7b monthly cost →| Model | Provider | Input /1M | Output /1M |
|---|---|---|---|
| Llama 3.2 3b | Meta | $0.020 | $0.020 |
| Deepseek R1 Distill Llama 8b | DeepSeek | $0.025 | $0.025 |
| Llama 3.2 1b | Meta | $0.027 | $0.201 |
| Qwen3 4b | Qwen | $0.030 | $0.030 |
| Qwen3 8b | Qwen | $0.035 | $0.138 |