Hollow markers & dashed spokes: axis not yet scored
G
60.5
A
—
U
—
N
—
T
n/d
L
—
E
—
T
3.7
Specification
- Parameters
- 32B
- Architecture
- qwen2
- Size on disk
- 34.82 GB
- Quantization
- 8bit
- Format
- MLX
- Runtime
- LM Studio
- Class
- Standard
- Reasoning (CoT)
- Yes — emits reasoning tokens
- Internal ID
- M12
- Mean speed
- 13.1 tok/s across suites
- Stall census
- 1 stall in 13 observed tests (7.7%)
- Reasoning appetite
- 1,057 tokens mean · 7,789 max
- Model card
- huggingface.co
Suite results
Suite Score Avg / 20 Tests n tok/s Run
General capability (13-task real-workload suite) 1a 157.3 / 260 12.1 13/13 n=2 13.1 run 2026-09-02
| Test | Category | Score | tok/s | |
|---|---|---|---|---|
| A1 | Dev/Ops Scripting | 9/20 | 12.9 | |
| A2 | Dev/Ops Scripting | 7/20 | 13.2 | |
| A3 | Dev/Ops Scripting | 19/20 | 13.4 | |
| A4 | Dev/Ops Scripting | 15/20 | 13.6 | |
| A5 | Dev/Ops Scripting | 10/20 | 13.7 | |
| A6 | Dev/Ops Scripting | 20/20 | 13.9 | |
| B1 | Document Processing | 20/20 | 13.5 | |
| B2 | Document Processing | 13/20 | 13.4 | |
| B3 | Document Processing | 0/20 | 11.2 | |
| C1 | Content Production | 14/20 | 12.3 | |
| C2 | Content Production | 10/20 | 12.6 | |
| C3 | Content Production | 9/20 | 12.8 | |
| C4 | Content Production | 11/20 | 13.3 |
Category mean Dev/Ops Scripting 13.3Document Processing 11Content Production 11
Each test is LLM-judged 0–20 against a fixed rubric; suite max = tests × 20. An ADJ chip marks a cell carrying a subjective adjustment from the maintainer; the judged score is the one shown, is unchanged, and remains what the axes, ranks and tie bands are computed from. Hover for the reason. Rows with a ▶ expand to the per-test breakdown. Where fewer tests ran than the suite total, unrun tests count as zero toward the axis score. See methodology.