gemma-4-26B-A4B-it

Gemma family · 27.0B params

4recorded runs

209.8best tok/s

2verified

Evidence: 2 sourced runs · 2 awaiting review · 1 independent source · first recorded Aug 16, 2026 · last recorded Aug 29, 2026. What these labels mean.

What the gemma-4-26B-A4B-it numbers mean

The fastest verified result on our board is 131.4 output tok/s on an Radeon AI Pro R9700 x2 at Q8_K_XL with a 262144-token context. That is about 7.6 ms per generated token — 500 tokens takes roughly 3.8 seconds. Prompt processing on the same record runs at 1974.7 tok/s, so a 2,000-token prompt is read in about 1.0 seconds. Across the 4 records we hold for this model the range is 43.3–209.8 tok/s on 4 GPUs, because quantisation, context length and the engine build all move the number.

How much VRAM does gemma-4-26B-A4B-it need?

gemma-4-26B-A4B-it has 27.0B parameters. These estimates use the same formula as the VRAM calculator: weights, plus about 15% for a moderate-context KV cache and runtime overhead, plus a small engine floor.

Quantisation	Estimated VRAM	Smallest card that fits
Q4_K_M (4-bit)	18.3 GB	24 GB
Q8_0 (8-bit)	33.8 GB	48 GB
FP16 (16-bit)	62.9 GB	80 GB

Long contexts and large batches raise the real number — read the smallest card as "possible with a reduced context". Measured throughput is on the benchmark board.

Fastest GPUs for this model (best tok/s)

Models with a similar footprint

Model	Params	Q4_K_M weights
gemma-4-26B-A4B-it-qat	27B	15.2 GB
gemma-4-26B-A4B-it-ultra-uncensored-heretic	27B	15.2 GB
Gemma4-26B-A4B-QAT-Uncensored-HauhauCS-Balanced-MTP	27B	15.2 GB
Ornstein3.6-27B-MTP-NSC-ACE-SABER	27B	15.2 GB
Qwen3.8-27B-MTP	27B	15.2 GB
Ternary Bonsai 2 27B	27B	15.2 GB

Ranked by distance from gemma-4-26B-A4B-it's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
Radeon AI Pro R9700 x2 · 32 GB	gemma-4-26B-A4B-it	Q8_K_XL	llama.cpp 5ea1b124e + stew675 01/10/02	262144	131.4 ✓	1974.7	—	src
RTX 3060 x2 · 12 GB	gemma-4-26B-A4B-it	UD-Q4_K_M	llama.cpp b10443	65536	62.1 ✓	794.6	18.0	src
RTX 5090 · 32 GB	gemma-4-26B-A4B-it	Q4_K_M	lmstudio —	2048	209.8 …	458.4	—	src
Intel Arc Pro B70 · 32 GB	gemma-4-26B-A4B-it	Q8_K_XL	llama.cpp —	2048	43.3 …	1303.4	—	src