Qwen · open weights · medium
Qwen: Qwen3 Coder 30B A3B Instruct
Open weights, so you can run it locally. The published id includes 30B-A3B. That label is copied from the id, not measured here. Size class is this site's filter bucket, not a benchmark.
API price, in pounds
- Input, per million
- £0.112
- Output, per million
- £0.447
OpenRouter lists $0.1500 in and $0.6000 out per million tokens. Converted at 0.74501 on 2026-11-10. Fetched 2026-11-11. Context length listed as 262,144 tokens.
Sourced local speeds
- 304.9 reply tokens/sec on GeForce RTX 5090, Q4_K_M. Source. Run r_vsype4pm1ko, marked Measured by llm-speed. Chat-short. No power draw in the row. Page read 11 November 2026.
Comparisons
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 5090
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 3090
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 4090
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 4070 Ti SUPER
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 4080 SUPER
- Qwen: Qwen3 Coder 30B A3B Instruct on GeForce RTX 5070 Ti