AMD
6 modelsProvider docs
As of Oct 4, 2026, AMD lists 6 supported models in the Sovyron catalog, including DeepSeek V4.1 Flash, MiniCPM5-2B, Qwen3.8 Flash Next, DeepSeek V4 Flash Vision Exp and Qwen3.8 27B — deepseek-flash and qwen families. Its metered rates span $0.175–$0.2786 per 1M tokens on a 3:1 input:output blend, and every price in the table is AMD's own, normalized to USD — open a model to compare it with the cheapest offer across all providers. Offer mix: 1 free-tier offer.
Supported models and pricing
6 of 6 models
| Model | 1M input | 1M output | 1M cache read | Context | Released | Tags |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | $0.14 | $0.28 | $0.0028 | 1.0M | Sep 10, 2026 | |
| MiniCPM5-2B | $0.124 | $0.7425 | $0.124 | 131k | Sep 6, 2026 | |
| Qwen3.8 Flash Next | $0.15 | $0.47 | $0.016 | 262k | Aug 27, 2026 | |
| DeepSeek V4 Flash Vision Exp | $0.14 | $0.28 | $0.0028 | 1.0M | Aug 21, 2026 | |
| Qwen3.8 27B | free | free | — | 131k | Aug 14, 2026 | free |
| DeepSeek V4 Flash 0731 | $0.14 | $0.28 | $0.0028 | 1.0M | Jul 31, 2026 |