- Parameters
- 4B
- Quant
- Q4_K_M
- Size
- 2.4 GB
- Family
- spark
spark-x2.5-4b
Measured, not seateddenseresidentllama.cpp (build 10893+)
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
Hybrid attention - one full-attention layer per three sliding-window layers - giving a native 1M context. Apache 2.0, 20T training tokens, marketed for agentic workflows with claimed native compatibility for Codex, Claude Code and Hermes.
What came of it
CORRECTED 2026-09-15 -- not the fastest on this board, and was never close. 48.3 tok/s short, 22.7 at 16k, 3.2 GB resident. Perfect on plain absence (24/24, matching gemma) and 95% on tool selection. But 0/60 on toolcall - it never perceives an unannounced tool - and worldview 20.0%, the lowest on the board. A dispatcher, not a router, and never a citer. Requires llama.cpp build 10893 or later for the spark2_5 architecture.
No runs recorded against this identifier.