Results / LFM2.5 24B A2B Q4_K_M (llama.cpp)
LFM2.5 24B A2B Q4_K_M (llama.cpp)
hermes-lfm25llamacppomarchy (RTX 3090 24 GB)scene
This run produced no playable build.
Metrics
- Checklist score
- – / 100
- Design (rubric, 0–5)
- –
- Stop reason
- done
- Model time
- 1min
- Tool time (wall minus model)
- 5s
- Wall time
- 2mins
- Tool calls
- 5
- Tool errors
- 3
- Errors by type
- {"other": 3}
- Loop-guard events
- 0
- Tokens in / out
- 65,802.0 / 17,296.0
- Thinking tokens
- –
- Decode / prefill tok/s
- 208.4 / 5,208.3
- Prefix reuse
- 0.765
- Peak VRAM (GPU / engine)
- 15,071 / 15,062 MiB
- Peak engine RAM
- 1,268 MiB
- Engine
- llamacpp
- Model
- LFM2.5 24B A2B Q4_K_M (llama.cpp) (lfm25-24b-a2b)
- Settings
- {"thinking": "medium", "reasoning_budget_tokens": 3072, "max_tokens": 7168, "sampling": "card: temp 0.1 top_k 50 repeat_penalty 1.05 (server)"}
- Node
- omarchy (RTX 3090 24 GB)
- Prompt
- scene-pagoda-v1
- Started
- 2026-10-30T01:11:53Z
Feature scores
Screenshots
Landing page at desktop and phone widths, and the game at noon, at night, after breaking a block, and on a phone.
Checklist (0 pass, 0 partial, 0 fail)
| Feature | Test | Score | Evidence |
|---|
Suite · · WebGL:
Steers
None: an unsteered run gets the prompt and nothing else.