Qwen3.5-27B

193 records across 21 hardware.

hardware	model	quant	backend	tok/s	pp tok/s	W	ttft s	ctx/tg	date	class	flags	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	60.9	3338.5	–	–	4096	–	sourced	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K	llama.cpp	58.8	3004.1	–	–	4096	–	sourced	–	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	57.5	2980.5	–	–	16384	–	sourced	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K	llama.cpp	55.8	2721.5	–	–	16384	–	sourced	–	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	55.1	2526.7	–	–	32768	–	sourced	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K	llama.cpp	53.8	2341.8	–	–	32768	–	sourced	–	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	51.3	1856.7	–	–	65536	–	sourced	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K	llama.cpp	50.1	1606.4	–	–	65536	–	sourced	–	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	45.1	1404.9	–	–	131072	–	sourced	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K	llama.cpp	44.1	1019.8	–	–	131072	–	sourced	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	38.0	322.7	–	–	1024	2026-03-08	community	–	source
M3 Ultra (256 GB)	Qwen3.5-27B	4bit	MLX	38	–	–	–	–	–	sourced	–	source
RTX Pro 6000 Blackwell	Qwen3.5-27B 27B	Q4_K	llama.cpp	36.3	903.1	–	–	262144	–	sourced	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	35.3	324.2	–	–	4096	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	33.6	319.7	–	–	8192	2026-03-08	community	–	source
RTX 3090	Qwen3.5-27B 27B	Q4_K	llama.cpp	33.5	1104.2	–	–	4096	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.5	226.5	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.5	226.5	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	33.4	226.5	–	–	1024	2026-03-08	community	–	source
RTX 3090	Qwen3.5-27B 27B	Q4_K	llama.cpp	32.3	977.4	–	–	16384	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	32.2	180.4	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	32.2	180.4	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	32.2	180.4	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	32.2	180.4	–	–	1024	2026-03-09	community	–	source
M5 Max	Qwen3.5-27B	4bit	oMLX	32.0	942.6	–	–	4096	2026-10-16	community	–	source
M5 Max (128 GB)	Qwen3.5-27B	4bit	MLX	31.6	–	–	–	–	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.5	228.4	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.5	228.4	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	31.3	228.9	–	–	4096	2026-03-08	community	–	source
M5 Max (48 GB)	Qwen3.5-27B	4bit	MLX	31.3	779	–	–	8000	–	sourced	–	source
RTX 3090	Qwen3.5-27B 27B	Q4_K	llama.cpp	31.0	848.2	–	–	32768	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	30.4	182.4	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	30.4	182.4	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	30.4	182.4	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	30.4	182.4	–	–	4096	2026-03-09	community	–	source
RTX 5090	Qwen3.5-27B 27B	Q4_K_M	llama.cpp	30.09	1515.83	–	–	1024	2026-09	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	30.0	310.0	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	29.3	227.0	–	–	8192	2026-03-08	community	–	source
RTX 3090	Qwen3.5-27B 27B	Q4_K	llama.cpp	28.8	678.9	–	–	65536	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	28.4	180.9	–	–	8192	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	28.4	180.9	–	–	8192	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	28.4	180.9	–	–	8192	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	28.4	180.9	–	–	8192	2026-03-09	community	–	source
M5 Max	Qwen3.5-27B	4bit	oMLX	28.1	611.8	–	–	1024	2026-10-16	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	27.6	239.4	–	–	1024	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	27.6	239.4	–	–	1024	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	27.6	239.4	–	–	1024	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	26.6	222.3	–	–	16384	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	25.3	176.2	–	–	16384	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	25.3	176.2	–	–	16384	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	25.3	176.2	–	–	16384	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	25.3	176.2	–	–	16384	2026-03-09	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	24.9	291.3	–	–	32768	2026-03-08	community	–	source
M5 Max (48 GB)	Qwen3.5-27B	Q4_K - Medium	llama.cpp	23.7	171	–	–	8000	–	sourced	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	23.6	199.2	–	–	4096	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	23.6	199.2	–	–	4096	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	23.6	199.2	–	–	4096	2026-03-09	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	21.9	212.9	–	–	32768	2026-03-08	community	–	source
M2 Ultra (GPU count not published, 128 GB)	Qwen3.5-27B	8bit	MLX	20.6	–	–	–	–	–	sourced	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	20.5	169.1	–	–	32768	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	6bit	oMLX	20.5	240.5	–	–	1024	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	6bit	oMLX	20.5	240.5	–	–	1024	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	20.3	186.4	–	–	8192	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	20.3	186.4	–	–	8192	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	4bit	oMLX	20.3	186.4	–	–	8192	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	6bit	oMLX	19.5	235.0	–	–	4096	2026-03-09	community	–	source
M4 Max	Qwen3.5-27B	6bit	oMLX	19.5	235.0	–	–	4096	2026-03-09	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M3 Ultra	Qwen3.5-27B	4bit	oMLX	18.1	258.7	–	–	65536	2026-03-08	community	–	source
M5 Pro	Qwen3.5-27B	4bit	oMLX	17.4	514.0	–	–	4096	2026-10-07	community	–	source
M5 Pro	Qwen3.5-27B	4bit	oMLX	17.3	483.3	–	–	1024	2026-10-07	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M2 Ultra	Qwen3.5-27B	4bit	oMLX	16.8	192.2	–	–	65536	2026-03-08	community	–	source
M5 Pro	Qwen3.5-27B	4bit	oMLX	16.7	449.8	–	–	16384	2026-10-07	community	–	source
M4 Max	Qwen3.5-27B	Q4_K	llama.cpp	16.69	222.23	–	–	2048	–	sourced	–	source
M5 Pro	Qwen3.5-27B	4bit	oMLX	16.5	492.4	–	–	8192	2026-10-07	community	–	source
M5 Max (128 GB)	Qwen3.5-27B	Q6_K	llama.cpp	16.5	–	–	–	8192	–	sourced	–	source
M4 Max	Qwen3.5-27B	Q4_K	llama.cpp	16.14	209.3	–	–	8192	–	sourced	–	source
M4 Max	Qwen3.5-27B	Q4_K	llama.cpp	15.75	195.44	–	–	16384	–	sourced	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	15.4	68.0	–	–	1024	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	15.4	68.0	–	–	1024	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.4	126.3	–	–	1024	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.2	125.6	–	–	1024	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.2	125.6	–	–	1024	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.2	125.6	–	–	1024	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.2	125.6	–	–	1024	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	15.2	82.0	–	–	1024	2026-10-06	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	15.2	125.6	–	–	1024	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	15.2	82.0	–	–	1024	2026-10-06	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	15.2	82.0	–	–	1024	2026-10-06	community	–	source
M1 Max (64 GB)	Qwen3.5-27B	4bit	MLX	15	67	–	–	8000	–	sourced	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.8	125.9	–	–	4096	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.8	125.9	–	–	4096	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.8	125.9	–	–	4096	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.8	125.9	–	–	4096	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	14.8	77.3	–	–	4096	2026-10-06	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.8	125.9	–	–	4096	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	14.8	77.3	–	–	4096	2026-10-06	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	14.8	77.3	–	–	4096	2026-10-06	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	14.4	68.0	–	–	4096	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	14.4	68.0	–	–	4096	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.2	124.3	–	–	8192	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.2	124.3	–	–	8192	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.2	124.3	–	–	8192	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.2	124.3	–	–	8192	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	14.2	124.3	–	–	8192	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	13.4	69.0	–	–	16384	2026-10-06	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	13.4	69.0	–	–	16384	2026-10-06	community	–	source
M1 Max	Qwen3.5-27B	4bit	oMLX	13.4	69.0	–	–	16384	2026-10-06	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	12.3	121.0	–	–	16384	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	12.3	121.0	–	–	16384	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	12.3	121.0	–	–	16384	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	12.3	121.0	–	–	16384	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	12.3	121.0	–	–	16384	2026-03-09	community	–	source
M1 Max (64 GB)	Qwen3.5-27B	Q6_K	llama.cpp	12	–	–	–	–	–	sourced	–	source
M2 Max (38-core GPU, 96 GB)	Qwen3.5-27B	8bit	MLX	12	–	–	–	–	–	sourced	–	source
M1 Max (64 GB)	Qwen3.5-27B	Q4_K - Medium	llama.cpp	11.5	–	–	–	–	–	sourced	–	source
M1 Max	Qwen3.5-27B	8bit	oMLX	11.1	85.8	–	–	1024	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	8bit	oMLX	11.1	85.8	–	–	1024	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	8bit	oMLX	10.6	89.2	–	–	4096	2026-03-09	community	–	source
M1 Max	Qwen3.5-27B	8bit	oMLX	10.6	89.2	–	–	4096	2026-03-09	community	–	source
M1 Max (64 GB)	Qwen3.5-27B	Q8_0	llama.cpp	10.5	–	–	–	–	–	sourced	–	source
M5 Pro (64 GB)	Qwen3.5-27B	Q4_K - Medium	llama.cpp	10	–	–	2.156	–	–	sourced	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	9.7	114.0	–	–	32768	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	9.7	114.0	–	–	32768	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	9.7	114.0	–	–	32768	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	9.7	114.0	–	–	32768	2026-03-09	community	–	source
M4 Pro	Qwen3.5-27B	4bit	oMLX	9.7	114.0	–	–	32768	2026-03-09	community	–	source
M4 Pro (48 GB)	Qwen3.5-27B	8bit	MLX	8.5	–	–	–	–	–	sourced	–	source
Radeon 8060S	Qwen3.5-27B	W8A8	vLLM	6.27	–	–	–	tg400	2026-09-28	community	–	source
Radeon 8060S	Qwen3.5-27B	W8A8	vLLM	5.12	–	–	–	tg400	2026-09-28	community	–	source
Radeon 8060S	Qwen3.5-27B	W8A8	vLLM	4.67	–	–	–	tg400	2026-09-28	community	–	source
M4 (16 GB)	Qwen3.5-27B	Q4_K - Medium	llama.cpp	0.0084	–	–	66.9331	–	–	sourced	–	source
DGX Spark	Qwen3.5-27B	i1-Q3_K_M	llama.cpp (CUDA)	–	94	–	–	–	2026-04-02	community	–	source
DGX Spark	Qwen3.5-27B	i1-Q3_K_M	llama.cpp (CUDA)	–	80	–	–	–	2026-04-02	community	–	source

← all models