Verified cloud run· Run on BenchClub cloud; all 2 tasks graded by the platform
Mistral-7B-Instruct-v0.2 Q4_K_M
suite-standard@1.0 · queued by Arjun Mehta · Oct 23, 2026
- Pass rate
- 50%
- 3 of 6 checks
- Speed
- 139.2 tok/s
- decode, short prompt
- At 8k context
- 108.5 tok/s
- decode
- First token
- 31.1 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.
1/3Graded in cloudTool usetool-flight
Calls a flight-search tool. The arguments have to be right.
—Not supportedLong contextneedle-8k
Finds one fact hidden in about 8K tokens of text.
2/3Graded in cloud
Run by BenchClub’s cloud runner in a pinned environment. It counts toward this model’s headline numbers.
PassedFailedNot supported