# Inference API prices

Inference lists 9 model(s) in the Sovyron catalog. Every price is Inference's own, in USD per 1M tokens.
Prices as of 2026-10-04.

| Model | Input $/1M | Output $/1M | Cache read $/1M | Context | Released |
| --- | --- | --- | --- | --- | --- |
| Google Gemma 3 | $0.15 | $0.3 | — | 125k | Jan 1, 2025 |
| Llama 3.1 8B Instruct | $0.025 | $0.025 | — | 16k | Jan 1, 2025 |
| Llama 3.2 11B Vision Instruct | $0.055 | $0.055 | — | 16k | Jan 1, 2025 |
| Llama 3.2 1B Instruct | $0.01 | $0.01 | — | 16k | Jan 1, 2025 |
| Llama 3.2 3B Instruct | $0.02 | $0.02 | — | 16k | Jan 1, 2025 |
| Mistral Nemo 12B Instruct | $0.038 | $0.1 | — | 16k | Jan 1, 2025 |
| Osmosis Structure 0.6B | $0.1 | $0.5 | — | 4k | Jan 1, 2025 |
| Qwen 2.5 7B Vision Instruct | $0.2 | $0.2 | — | 125k | Jan 1, 2025 |
| Qwen 3 Embedding 4B | $0.01 | free | — | 32k | Jan 1, 2025 |

HTML page: https://sovyron.com/providers/inference/
Full dataset: https://sovyron.com/data/catalog.json