● FitLLM — fit receipt · 8-bit · 8K tokens ctx · KV F16 · engine 2.16.0 · revision f1e3678550d69a40971615209e14e399144b2272 · release manifest

GLM-4.7-Flash on Mac 32GB

WON'T FIT ✗

weights	KV cache	overhead	reserve	total	short by
29.1 GB	0.4 GB	3.8 GB	8.5 GB	41.8 / 32 GB	9.8 GB

→ Quantize to 4-bit and it fits (small quality cost).

Active structural premises

mla-compressed-latent-cache — KV memory assumes a compressed-latent MLA artifact or mode; legacy non-MLA GGUF or an explicitly uncompressed mode invalidates this estimate.

Replay: npx fitllm-engine "GLM-4.7-Flash" --mac 32 · JSON · interactive

live fit badge — embed: ![fits](https://img.shields.io/endpoint?url=https%3A%2F%2Ffitllm.run%2Fapi%2Fbadge%3Fmodel%3DGLM-4.7-Flash%26ram%3D32%26quant%3D8%26ctx%3D8192%26kv%3D16)

Computed by the open fitllm-engine (MIT) from official config.json values — estimates; runtime varies. Ran it for real? Challenge this prediction — measured reports calibrate the engine for everyone.