Hollow markers & dashed spokes: axis not yet scored
G
73.5
A
—
U
—
N
—
T
n/d
L
—
E
—
T
10.2
Specification
- Parameters
- 24B
- Architecture
- mistral3
- Size on disk
- 14.12 GB
- Quantization
- 4bit
- Format
- MLX
- Runtime
- LM Studio
- Class
- Standard
- Reasoning (CoT)
- No
- Internal ID
- M3
- Mean speed
- 35.8 tok/s across suites
- Stall census
- 0 stalls in 13 observed tests (0.0%)
- Model card
- huggingface.co
Suite results
Suite Score Avg / 20 Tests n tok/s Run
General capability (13-task real-workload suite) 1a 191.1 / 260 14.7 13/13 n=2.1 35.8 run 2026-09-02
| Test | Category | Score | tok/s | |
|---|---|---|---|---|
| A1 | Dev/Ops Scripting | 14/20 | 33 | |
| A2 | Dev/Ops Scripting | 13/20 | 31.7 | |
| A3 | Dev/Ops Scripting | 9/20 | 33 | |
| A4 | Dev/Ops Scripting | 19/20 | 33.2 | |
| A5 | Dev/Ops Scripting | 15/20 | 33.4 | |
| A6 | Dev/Ops Scripting | 20/20 | 35.2 | |
| B1 | Document Processing | 19/20 | 35.7 | |
| B2 | Document Processing | 20/20 | 35.7 | |
| B3 | Document Processing | 11/20 | 36.4 | |
| C1 | Content Production | 15/20 | 38.1 | |
| C2 | Content Production | 8/20 | 38 | |
| C3 | Content Production | 16/20 | 30.9 | |
| C4 | Content Production | 12/20 | 31.2 |
Category mean Dev/Ops Scripting 15Document Processing 16.7Content Production 12.8
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.