Qwen3-Coder-30B-A3B-Instruct
Qwen
BenchClub cloud
- Pass rate
- 89%
- 24 of 27 checks
- Speed
- 186.4 tok/s
- decode, short prompt
- At 8k context
- 160.3 tok/s
- decode
- Cloud runs
- 3
- speeds within 1.8%
- Coding9/9100%
- Tool use7/978%
- Long context8/989%
- Suite
- [email protected]
- Machine
- NVIDIA L40S · 48 GB
- Engine
- ollama 0.40.14
Headline numbers come only from cloud runs. Runs on members’ own machines are listed below and don’t change them.
LMArena
Not ranked on LMArena
LMArena only ranks the models it serves.
BenchClub runs
Run this modelRun byWhereSpeed
Saoirse ByrneNov 6, 2026RTX 4090 · 24 GB224.4 tok/s
- Niko LaineNov 4, 2026RTX 5090 · 32 GB371.7 tok/s
Samir HaddadNov 4, 2026L40S · 48 GBCloud186.4 tok/s
Ben CarterOct 30, 2026RX 7900 XTX · 24 GB162.3 tok/s
- Sofía RuizOct 15, 2026GB10 · 128 GB56.7 tok/s
- Niko LaineOct 6, 2026L40S · 48 GBCloud185.6 tok/s
- Kwame MensahOct 4, 2026L40S · 48 GBCloud188.9 tok/s
PassedFailedNot supported