- Parameters
- 8B / 1B active
- Quant
- Q4_K_M
- Size
- 5.0 GB
- Family
- lfm
RecordStatus · runtime · seat
lfm2.5-8b-a1b
Measured, not seatedMoEresidentllama.cpp
Model card ↗Outcome as of 20 September 2026, from the seat registry. The write-up below was made when the model was tested.
Why it was tested
Tested as an extremely sparse small MoE, 1B active, to find the floor of useful.
What came of it
Measured on grounding. Its main contribution was exposing a harness bug: a 16-token gate probe starved it and two other reasoning models, producing false empty-content alarms.
Recorded runs3 scored · 95% intervals
| Suite | Score | 95% interval | Passed | Runtime · profile | Recorded |
|---|---|---|---|---|---|
| grounding | 47.2% | 38.7–55.9 | Passed 59/125 | llama.cppextraction | 2026-08-31 |
| nullschema | 27.8% | 18.8–39.0 | Passed 20/72 | llama.cppextraction | 2026-08-31 |
| screening | 73.3% | 61.0–82.9 | Passed 44/60 | llama.cppextraction | 2026-08-31 |
Runs marked unenforced were measured through mlx_lm on a suite that requires constrained decoding. mlx_lm accepts the schema and ignores it, so those numbers do not measure what the suite is designed to measure. They are kept and labelled rather than deleted.
Every run log entry for this model (3)
| Finished | Suite | Runtime · profile | Score | Coverage | Grader · dataset · config |
|---|---|---|---|---|---|
| nullschemanullschema/1 | llama.cppextraction | 27.8% | Coverage 72/72 | g d7448e0f5fc3f9afd 3b91280bb40372efc 00612fdd4a39faf4 | |
| groundinggrounding/1 | llama.cppextraction | 47.2% | Coverage 125/125 | g 30b933dcc6a7e2d4d dcb396faf8b83b0ac 90be8dffe935ab41 | |
| screeningscreening/1 | llama.cppextraction | 73.3% | Coverage 60/60 | g 0661b49a10be2d2ad 64c97ca2269c29afc 2cbf094131c24516 |