Gemma 4 26B (4B active)
on NVIDIA RTX 4090 · llama.cpp · 262,144 ctx
Sep 26, 2026
Use cases
codingvisionagenticlong-context
Summary
User reports Gemma 4 26B-A4B at ~129 tok/s single-request generation on a single RTX 4090, with ~5,600 tok/s prefill.
Setup is a TurboQuant fork of llama.cpp with Q5_K_M weights, turbo3 3-bit KV cache, Flash Attention, and a multimodal projector, using 22.3 GB of 24 GB VRAM at 262,144 tokens of context.
Concurrent generation is reported at ~83-85 tok/s. A 15-test agentic coding benchmark through Claude Code passed 15/15 with an average quality of 4.2/5.