Qwen: Qwen3.8 27B
Qwen
Qwen: Qwen3.8 27B by Qwen costs $0.04 per 1M input tokens and $2.30 per 1M output tokens, with a 1M-token context window.
Pricing
Input (per 1M tokens)
$0.04
Output (per 1M tokens)
$2.30
Cached input (per 1M)
$0.02
Specifications
- Provider
- Qwen
- Context window
- 1M tokens
- Parameters
- —
- Released
- —
- Open weights
- —
- Frontier model
- No
Benchmarks
- LMArena Vision Elovision
- 1271.4
- LMArena (Chatbot Arena) Elohuman-preference
- 1440.6
- Epoch Capabilities Indexcomposite
- 149.4
- LMArena WebDev Elocoding
- 1594.2
Speed & latency
Time to first token
2.92s
Output speed
62 tok/s
End-to-end response
—
Compare Qwen: Qwen3.8 27B with…
Qwen: Qwen3.8 27B vs Qwen: Qwen3 VL 8B Thinking$1.37/1MQwen: Qwen3.8 27B vs Qwen: Qwen3.6 Max Preview$6.24/1MQwen: Qwen3.8 27B vs Qwen: Qwen3 VL 32B Instruct$0.42/1MQwen: Qwen3.8 27B vs Qwen: Qwen3 VL 235B A22B Thinking$2.60/1MQwen: Qwen3.8 27B vs Qwen: Qwen3 Coder Plus$3.25/1MQwen: Qwen3.8 27B vs Qwen: Qwen3 Next 80B A3B Instruct (free)$0.00/1M
FAQ
Pricing is per 1M tokens (USD); confirm with the provider before production use.