DeepInfra pricing overview
TokenWatch tracks 139 text, 0 image, and 0 video model records for this provider. TokenWatch does not have a provider-wide zero-retention verdict for this page.
Cheapest input price: Llama 3.1 8B Instruct at $0.020 per million input tokens and $0.040 per million output tokens. 64 of 139 text models publish a cache-read price; 69 carry a zero-data-retention tag.
DeepInfra text-model pricing
Text offerings ranked by the Agentic workload mix (2.5% input, 97% cached input, 0.5% output). Change the calculator mix for a workload-specific estimate.
| Org | Provider | Model | Input $/M | Output $/M | Cache $/M | Blended $/M |
|---|---|---|---|---|---|---|
| deepseek | deepinfra | DeepSeek: DeepSeek V4.1 Flash 1017 (flex) | $0.078 | $0.235 | $0.0024 | $0.0054 |
| xiaomi | deepinfra | XiaomiMiMo/MiMo-V2.6-Flash | $0.140 | $0.280 | $0.0028 | $0.0076 |
| deepseek | deepinfra | DeepSeek: DeepSeek V4.1 Flash (flex) | $0.112 | $0.336 | $0.0034 | $0.0077 |
| deepseek | deepinfra | DeepSeek: DeepSeek V4.1 Flash (batch) | $0.112 | $0.336 | $0.0034 | $0.0077 |
| deepseek | deepinfra | DeepSeek: DeepSeek V4.1 Flash 1017 (priority) | $0.147 | $0.441 | $0.0044 | $0.010 |
| deepseek | deepinfra | deepseek-ai/DeepSeek-V4.1-Flash-1017 | $0.200 | $0.600 | $0.0060 | $0.014 |
| deepseek | deepinfra | deepseek-ai/DeepSeek-V4.1-Flash | $0.200 | $0.600 | $0.0060 | $0.014 |
| inclusionai | deepinfra | inclusionAI/Ling-3.0-flash-VL | $0.060 | $0.180 | $0.012 | $0.014 |
| inclusionai | deepinfra | inclusionAI/Ling-3.0-flash | $0.060 | $0.180 | $0.012 | $0.014 |
| deepseek | deepinfra | DeepSeek: DeepSeek V4.1 Flash (priority) | $0.210 | $0.630 | $0.0063 | $0.015 |
| deepseek | deepinfra | deepseek-ai/DeepSeek-V4-Flash-0731 | $0.060 | $0.180 | $0.015 | $0.017 |
| ibm | deepinfra | ibm-granite/granite-4.2-8b | $0.060 | $0.250 | $0.015 | $0.017 |
| qwen | deepinfra | Qwen/Qwen3.8-Flash | $0.113 | $0.382 | $0.014 | $0.018 |
| xiaomi | deepinfra | XiaomiMiMo/MiMo-V2.6-Pro | $0.430 | $0.870 | $0.0036 | $0.019 |
| meta | deepinfra | meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo | $0.020 | $0.040 | — | $0.020 |
| meta | deepinfra | Meta: Llama 3.1 8B Instruct | $0.020 | $0.040 | — | $0.020 |
| deepseek | deepinfra | deepseek-ai/DeepSeek-V4-Flash | $0.090 | $0.180 | $0.018 | $0.021 |
| bytedance | deepinfra | ByteDance/Seed-2.0-mini | $0.100 | $0.400 | $0.020 | $0.024 |
| openai | deepinfra | OpenAI: gpt-oss-20b (batch) | $0.024 | $0.112 | — | $0.024 |
| nvidia | deepinfra | nvidia/NVIDIA-Nemotron-3.5-Super-VL | $0.150 | $0.500 | $0.020 | $0.026 |
| nvidia | deepinfra | NVIDIA: Nemotron 3.5 Super VL | $0.150 | $0.500 | $0.020 | $0.026 |
| z-ai | deepinfra | Z.ai: GLM 5.3 Flash (batch) | $0.120 | $0.400 | $0.024 | $0.028 |
| mistral | deepinfra | mistralai/Mistral-Nemo-Instruct-2407 | $0.029 | $0.030 | — | $0.029 |
| mistral | deepinfra | Mistral: Mistral Nemo | $0.029 | $0.030 | — | $0.029 |
| openai | deepinfra | OpenAI: gpt-oss-120b (batch) | $0.030 | $0.136 | — | $0.030 |
Catalog snapshot 2026-10-29 from direct-provider APIs, OpenRouter endpoint data, and maintained fallbacks. See the sources and calculation rules; verify rates on the provider's official pricing page before committing spend.