CMP 170HX

Nvidia · 64 GB VRAM

23recorded runs

1392.4best tok/s

16sourced

How it compares — 64 GB tier

CMP 170HX255.9 Mac Studio (M4 Max)226.9 Mac Studio (Mac16,9) with Apple M4 Max110.3 CMP 170HX x1029.7

Best verified tok/s per GPU in this VRAM tier — compare any two like-for-like.

Evidence: 16 sourced runs · 7 awaiting review · 2 independent sources · first recorded Aug 27, 2026 · last recorded Sep 6, 2026. What these labels mean.

What fits on 64 GB — measured

Model	Quant	Peak VRAM	Headroom
gemma-4-31B-it	NVFP4	58.7 GB	5.3 GB
Llama-3.3-70B-Instruct	NVFP4	57.4 GB	6.6 GB
Llama-4-Scout-17B-16E-Instruct	UD-Q3_K_XL	46.6 GB	17.4 GB
Qwen3.8-Flash-Next	UD-IQ1_S	43.2 GB	20.8 GB

Peak process memory recorded by the submitter on this card — evidence of what fit, not a guarantee. Plan with the VRAM calculator.

All benchmarks Compare CMP 170HX →

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
CMP 170HX · 64 GB	Qwen3.8-27B	W4A16	vllm 0.27.1	4359	255.9 ✓	2321.6	—	src
CMP 170HX · 64 GB	Qwen3.8-27B	BF16	vllm 0.27.1	4352	181.6 ✓	—	—	src
CMP 170HX · 64 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	Q8_0	llama.cpp 0.3.0-dev+b10705	67592	165.1 ✓	1524.7	—	src
CMP 170HX · 64 GB	Qwen3.8-27B	INT8	vllm 0.27.1	65536	147.7 ✓	—	—	src
CMP 170HX · 64 GB	Qwen3.8-27B	W4A16	vllm 0.27.1	65536	140.5 ✓	—	—	src
CMP 170HX · 64 GB	Qwen3.6-35B-A3B	UD-Q8_K_XL	llama.cpp 0.3.0-dev+b10705	131338	87.2 ✓	1322.2	—	src
CMP 170HX · 64 GB	Ornith-1.5-35B-A3B	Q8_0	llama.cpp 0.3.0-dev+b10705	131338	87.1 ✓	1329.4	—	src
CMP 170HX · 64 GB	Ornith-1.5-9B	BF16	vllm 0.27.1	131334	61.5 ✓	5305.1	—	src
CMP 170HX · 64 GB	Ornith-1.5-35B-A3B	W4A16	vllm 0.27.1	131335	48.6 ✓	5318.3	—	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-Flash-Next	UD-IQ1_S	llama.cpp pr-27742	65536	46.1 ✓	68.7	43.2	src
CMP 170HX · 64 GB	Llama-4-Scout-17B-16E-Instruct	UD-Q3_K_XL	llama.cpp 0.28.0	2048	44.5 ✓	584.4	46.6	src
CMP 170HX · 64 GB	gemma-4-31B-it	NVFP4	vllm 0.28.0	32768	28.3 ✓	967.9	58.7	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
CMP 170HX · 64 GB	granite-4.2-30b	Q8_0	llama.cpp 0.3.0-dev+b10705	65797	23.7 ✓	624.0	—	src
CMP 170HX · 64 GB	Llama-3.3-70B-Instruct	NVFP4	vllm 0.28.0	32768	20.1 ✓	626.2	57.4	src
CMP 170HX · 64 GB	Qwen3.8-27B	W4A16	sglang 0.5.19	1280	1392.4 …	—	—	src
CMP 170HX · 64 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	W4A16	vllm 0.28.0	67592	217.9 …	8392.9	—	src
CMP 170HX · 64 GB	Qwen3.6-35B-A3B	FP8	vllm 0.28.0	131335	62.0 …	4982.8	—	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
CMP 170HX · 64 GB	Ornith-1.5-35B-A3B-MTP	FP8	vllm 0.27.1	131334	33.2 …	4984.8	—	src
CMP 170HX · 64 GB	Qwen3.8-27B	UD-Q4_K_XL	llama.cpp 0.3.0-dev+b10705	131295	29.7 …	595.7	—	src
CMP 170HX · 64 GB	Qwen3.8-27B	UD-Q8_K_XL	llama.cpp 0.3.0-dev+b10705	131295	25.9 …	610.2	—	src

Related reading: GPU buying guide · used GPUs · Value index · Street prices