Mac Studio, M5 Ultra, 512GB

1TB of storage, 1.2TB/s of memory bandwidth, and 504GB of unified memory left for a model once macOS has taken its share. It holds 28 of the 30 models we track.

Offers

no price yet

Ships late October 2026. Apple has not published a price.

Fit and Speed below are computed from published specs, so they are already true.

Timing

Collecting price history — 0 of 90 days. No timing verdict yet.

Models it runs — 8K context

Llama-4-Behemoth—
Won't fit— headroom
DeepSeek-V4-Pro—
Won't fit— headroom
Kimi-K2-InstructQ3_K_M
Tight13GB headroomest. 38.4 tok/s
DeepSeek-V3Q5_K_M
Tight27GB headroomest. 22.9 tok/s
DeepSeek-R1Q5_K_M
Tight27GB headroomest. 22.9 tok/s
Qwen3-Coder-480B-A35B-InstructQ6_K
Fits107GB headroomest. 20.9 tok/s
Llama-4-Maverick-17B-128EQ8_0
Fits76GB headroomest. 33.2 tok/s
GLM-4.6Q8_0
Fits120GB headroomest. 17.6 tok/s
DeepSeek-V4-FlashQ8_0
Fits201GB headroomest. 43.4 tok/s
Qwen3-235B-A22BBF16
Tight31GB headroomest. 13.6 tok/s
MiniMax-M2BF16
Fits43GB headroomest. 30.0 tok/s
Mistral-Large-2411BF16
Fits254GB headroomest. 3.2 tok/s
gpt-oss-120bBF16
Fits268GB headroomest. 58.8 tok/s
Llama-4-Scout-17B-16EBF16
Fits283GB headroomest. 17.6 tok/s
Qwen3-Coder-NextBF16
Fits343GB headroomest. 100 tok/s
Llama-3.3-70BBF16
Fits360GB headroomest. 5.6 tok/s
Qwen3.5-35B-A3BBF16
Fits433GB headroomest. 100 tok/s
Qwen3-32BBF16
Fits437GB headroomest. 12.2 tok/s
Gemma-4-31BBF16
Fits440GB headroomest. 12.7 tok/s
Qwen3-30B-A3BBF16
Fits441GB headroomest. 90.9 tok/s
Gemma-3-27BBF16
Fits447GB headroomest. 14.4 tok/s
Qwen3.6-27BBF16
Fits448GB headroomest. 14.4 tok/s
Gemma-4-26B-A4BBF16
Fits452GB headroomest. 78.9 tok/s
Devstral-Small-2507BF16
Fits453GB headroomest. 16.3 tok/s
gpt-oss-20bBF16
Fits460GB headroomest. 83.3 tok/s
Phi-4BF16
Fits473GB headroomest. 27.9 tok/s
Gemma-4-12BBF16
Fits478GB headroomest. 32.6 tok/s
Qwen3-VL-8B-InstructBF16
Fits486GB headroomest. 48.8 tok/s
Gemma-3n-E4BBF16
Fits489GB headroomest. 97.5 tok/s
Qwen3-4BBF16
Fits494GB headroomest. 97.5 tok/s

Model	Best quant	Fit	Headroom	Est. speed
Llama-4-Behemoth	—	Won't fit	—	—
DeepSeek-V4-Pro	—	Won't fit	—	—
Kimi-K2-Instruct	Q3_K_M	Tight	13GB	38.4 tok/s
DeepSeek-V3	Q5_K_M	Tight	27GB	22.9 tok/s
DeepSeek-R1	Q5_K_M	Tight	27GB	22.9 tok/s
Qwen3-Coder-480B-A35B-Instruct	Q6_K	Fits	107GB	20.9 tok/s
Llama-4-Maverick-17B-128E	Q8_0	Fits	76GB	33.2 tok/s
GLM-4.6	Q8_0	Fits	120GB	17.6 tok/s
DeepSeek-V4-Flash	Q8_0	Fits	201GB	43.4 tok/s
Qwen3-235B-A22B	BF16	Tight	31GB	13.6 tok/s
MiniMax-M2	BF16	Fits	43GB	30.0 tok/s
Mistral-Large-2411	BF16	Fits	254GB	3.2 tok/s
gpt-oss-120b	BF16	Fits	268GB	58.8 tok/s
Llama-4-Scout-17B-16E	BF16	Fits	283GB	17.6 tok/s
Qwen3-Coder-Next	BF16	Fits	343GB	100 tok/s
Llama-3.3-70B	BF16	Fits	360GB	5.6 tok/s
Qwen3.5-35B-A3B	BF16	Fits	433GB	100 tok/s
Qwen3-32B	BF16	Fits	437GB	12.2 tok/s
Gemma-4-31B	BF16	Fits	440GB	12.7 tok/s
Qwen3-30B-A3B	BF16	Fits	441GB	90.9 tok/s
Gemma-3-27B	BF16	Fits	447GB	14.4 tok/s
Qwen3.6-27B	BF16	Fits	448GB	14.4 tok/s
Gemma-4-26B-A4B	BF16	Fits	452GB	78.9 tok/s
Devstral-Small-2507	BF16	Fits	453GB	16.3 tok/s
gpt-oss-20b	BF16	Fits	460GB	83.3 tok/s
Phi-4	BF16	Fits	473GB	27.9 tok/s
Gemma-4-12B	BF16	Fits	478GB	32.6 tok/s
Qwen3-VL-8B-Instruct	BF16	Fits	486GB	48.8 tok/s
Gemma-3n-E4B	BF16	Fits	489GB	97.5 tok/s
Qwen3-4B	BF16	Fits	494GB	97.5 tok/s

Best quant is the highest-quality one that still loads with headroom. Every speed is an estimate; the formula is here.