Hollow markers & dashed spokes: axis not yet scored
G
86
A
96
U
—
N
—
T
100
L
—
E
83.5
T
32.6
Specification
- Parameters
- 118B total / 8B active (MoE, 10-of-256 experts + 1 shared)
- Architecture
- laguna
- Size on disk
- 71.2 GB
- Quantization
- Q4_K_M
- Format
- GGUF
- Reasoning (CoT)
- No
- Internal ID
- M20
- Mean speed
- 53.3 tok/s across suites
- Stall census
- 0 stalls in 29 observed tests (0.0%)
- Model card
- huggingface.co
Suite results
| Suite | Score | Avg / 20 | Tests | tok/s |
|---|---|---|---|---|
| General capability (13-task real-workload suite) 1a | 223.6 / 260 | 17.2 | 13/13 | 59.5 |
| Agentic tool-calling & protocol adherence 1b | 153.6 / 160 | 19.2 | 8/8 | 37.4 |
| Coding depth 1c | 100.2 / 120 | 16.7 | 6/6 | 63 |
Each test is LLM-judged 0–20 against a fixed rubric; suite max = tests × 20. Where fewer tests ran than the suite total, unrun tests count as zero toward the axis score. See methodology.