IO.NET
17 modelsProvider docs
As of Oct 4, 2026, IO.NET lists 17 supported models in the Sovyron catalog, including Qwen 3 235B Thinking, Magistral Small 2506, Devstral Small 2505, DeepSeek-R1 and Llama 4 Maverick 17B 128E Instruct — qwen, magistral-small, devstral, deepseek-thinking, llama and gpt-oss families. Its metered rates span $0.025–$3.6875 per 1M tokens on a 3:1 input:output blend, and every price in the table is IO.NET's own, normalized to USD — open a model to compare it with the cheapest offer across all providers.
Supported models and pricing
17 of 17 models
| Model | 1M input | 1M output | 1M cache read | Context | Released |
|---|---|---|---|---|---|
| Qwen 3 235B Thinking | $0.11 | $0.6 | $0.055 | 262k | Jul 1, 2025 |
| Magistral Small 2506 | $0.5 | $1.5 | $0.25 | 128k | Jun 1, 2025 |
| Devstral Small 2505 | $0.05 | $0.22 | $0.025 | 128k | May 1, 2025 |
| DeepSeek-R1 | $2 | $8.75 | $1 | 128k | Jan 20, 2025 |
| Llama 4 Maverick 17B 128E Instruct | $0.15 | $0.6 | $0.075 | 430k | Jan 15, 2025 |
| Qwen 3 Coder 480B | $0.22 | $0.95 | $0.11 | 106k | Jan 15, 2025 |
| Qwen 3 Next 80B Instruct | $0.1 | $0.8 | $0.05 | 262k | Jan 10, 2025 |
| Llama-3.3-70B-Instruct | $0.13 | $0.38 | $0.065 | 128k | Dec 6, 2024 |
| GPT OSS 120B | $0.04 | $0.4 | $0.02 | 131k | Dec 1, 2024 |
| GPT OSS 20B | $0.03 | $0.14 | $0.015 | 64k | Dec 1, 2024 |
| GLM-4.6 | $0.4 | $1.75 | $0.2 | 200k | Nov 15, 2024 |
| Kimi K2 Thinking | $0.55 | $2.25 | $0.275 | 33k | Nov 1, 2024 |
| Mistral Large Instruct 2411 | $2 | $6 | $1 | 128k | Nov 1, 2024 |
| Qwen 2.5 VL 32B Instruct | $0.05 | $0.22 | $0.025 | 32k | Nov 1, 2024 |
| Llama 3.2 90B Vision Instruct | $0.35 | $0.4 | $0.175 | 16k | Sep 25, 2024 |
| Kimi K2 Instruct | $0.39 | $1.9 | $0.195 | 33k | Sep 5, 2024 |
| Mistral Nemo Instruct 2407 | $0.02 | $0.04 | $0.01 | 128k | Jul 1, 2024 |