Verified cloud run· Run on BenchClub cloud; all 3 tasks graded by the platform
Qwen3-30B-A3B-Instruct-2507 Q4_K_M
suite-standard@1.0 · queued by Ben Carter · Oct 26, 2026
- Pass rate
- 89%
- 8 of 9 checks
- Speed
- 189.4 tok/s
- decode, short prompt
- At 8k context
- 162.8 tok/s
- decode
- First token
- 34.1 ms
- short prompt
Tasks
Each task runs 3 times and is graded by a program, not by another model.
Codingcode-intervals
Writes a Python function. Unit tests decide the pass.
3/3Graded in cloudTool usetool-flight
Calls a flight-search tool. The arguments have to be right.
2/3Graded in cloudLong contextneedle-8k
Finds one fact hidden in about 8K tokens of text.
3/3Graded in cloud
Run by BenchClub’s cloud runner in a pinned environment. It counts toward this model’s headline numbers.
PassedFailedNot supported