Qwen3.8-27B-MTP

Qwen family · 27.0B params

7recorded runs

117.7best tok/s

3verified

Evidence: 3 sourced runs · 4 awaiting review · 1 independent source · first recorded Aug 15, 2026 · last recorded Sep 4, 2026. What these labels mean.

What the Qwen3.8-27B-MTP numbers mean

The fastest verified result on our board is 112.1 output tok/s on an RTX PRO 6000 Blackwell at NVFP4 with a 2048-token context. That is about 8.9 ms per generated token — 500 tokens takes roughly 4.5 seconds. Across the 7 records we hold for this model the range is 27.4–117.7 tok/s on 5 GPUs, because quantisation, context length and the engine build all move the number.

How much VRAM does Qwen3.8-27B-MTP need?

Qwen3.8-27B-MTP 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	27B	15.2 GB
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
Ternary Bonsai 2 27B	27B	15.2 GB

Ranked by distance from Qwen3.8-27B-MTP's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX PRO 6000 Blackwell · 96 GB	Qwen3.8-27B-MTP	NVFP4	vllm 0.27.1	2048	112.1 ✓	—	—	src
CMP 170HX · 64 GB	Qwen3.8-27B-MTP	INT8-W8A16	vllm 0.27.1	131336	46.4 ✓	1323.5	—	src
CMP 170HX · 64 GB	Qwen3.8-27B-MTP	MTP-Q8_0	llama.cpp 0.3.0-dev+b10705	131295	27.4 ✓	598.5	—	src
RTX 5090 · 32 GB	Qwen3.8-27B-MTP	NVFP4	vllm —	172800	117.7 …	7466.2	—	src
Intel Arc Pro B70 · 32 GB	Qwen3.8-27B-MTP	GPTQ	vllm —	2048	60.9 …	456.7	—	src
CMP 170HX · 64 GB	Qwen3.8-27B-MTP	MTP-Q4_K_M	llama.cpp 0.3.0-dev+b10705	17370	57.6 …	671.2	—	src
Tesla V100 x2 · 32 GB	Qwen3.8-27B-MTP	Q6_K	llama.cpp —	262144	38.3 …	282.1	—	src