Same open weights, served by 2 providers. Weights: Qwen/Qwen3.6-27B.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra | Qwen/Qwen3.6-27B | flexFP80.8× standard | $0.256 | $2.56 | — | $0.832 | Verified 4h ago · Changed 7h ago | |
DeepInfra | Qwen/Qwen3.6-27B | standardFP8 | $0.32 | $3.2 | — | $1.04 | Verified 4h ago · Changed 7h ago | |
Groq | qwen/qwen3.6-27b | standardQuantization not disclosedReduced context (131k of 262k) | $0.60 | $3.0 | $0.30 | $1.2 | Verified 6h ago · Changed 6h ago | |
DeepInfra | Qwen/Qwen3.6-27B | priorityFP81.5× standard | $0.48 | $4.8 | — | $1.56 | Verified 4h ago · Changed 7h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra fp8 | $96/mo |
| Groq | $120/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.