APPLE· M4 ULTRA

Apple M4 Ultra (384GB)

VRAM
384 GB
FLAGSHIP
BANDWIDTH
1092
GB/S
MODELS Q4
411/428
96%
7B Q4 SPEED
~139
BLAZING
▸ MODEL COVERAGE @ Q496% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

Average speeds at Q4 quantization. Actual performance varies by model architecture and context length.

3B
~324
TOK/S
7B
~139
TOK/S
14B
~69
TOK/S
32B
~30
TOK/S
70B
~14
TOK/S
▸ SPECIFICATIONS
VRAM
384 GB
BANDWIDTH
1092 GB/s
FP16 COMPUTE
40 TFLOPS
TDP
215W
MEMORY
LPDDR5X
ARCHITECTURE
M4 Ultra
MSRP
$9,999
341
FAST MODELS · >30 TOK/S
Real-time chat speed
394
USABLE · >10 TOK/S
Comfortable for all tasks
411
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test M4 Ultra (384GB) (or anything bigger) without committing. Pay by the second, cancel anytime.

Spin up in ~60s. Pay by the second. Cancel anytime.

Some links are affiliate links — we may earn a small commission at no extra cost to you. This helps keep FitMyLLM free and independent.

▸ COMPATIBLE MODELS· 411
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
10785
TOK/S · 0% VRAM
S
nomic-embed-text-v1.5 100M0.1B
EMBEDDING·8K CTX· CHAT
9707
TOK/S · 0% VRAM
S
GPT-2 124M0.124B
GPT2·1K CTX· CHAT
7828
TOK/S · 0% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
7190
TOK/S · 0% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
3792
TOK/S · 0% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
3595
TOK/S · 0% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
3183
TOK/S · 0% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
2898
TOK/S · 0% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
2898
TOK/S · 0% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
2898
TOK/S · 0% VRAM
S
GPT-2 Medium 345M0.345B
GPT2·1K CTX· CHAT
2814
TOK/S · 0% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
2773
TOK/S · 0% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
2696
TOK/S · 0% VRAM
S
Falcon-H1 0.5B0.5B
FALCON·128K CTX· CHAT
1941
TOK/S · 0% VRAM
S
Qwen 1.5 0.5B0.5B
QWEN·32K CTX· CHAT
1941
TOK/S · 0% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
1941
TOK/S · 0% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
1941
TOK/S · 0% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
1709
TOK/S · 0% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
1709
TOK/S · 0% VRAM
S
Qwen3 0.6B0.6B
QWEN·32K CTX· CHAT· REASONING
1618
TOK/S · 0% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
1618
TOK/S · 0% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
1618
TOK/S · 0% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
1618
TOK/S · 0% VRAM
S
LFM2 700M0.7B
LFM·125K CTX· CHAT· TOOL_USE
1387
TOK/S · 0% VRAM
S
GPT-2 Large 774M0.774B
GPT2·1K CTX· CHAT
1254
TOK/S · 0% VRAM
S
Qwen 3.5 0.8B0.8B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
1213
TOK/S · 0% VRAM
S
Qwen3.5-0.8B0.9B
QWEN·256K CTX· CHAT
1079
TOK/S · 0% VRAM
S
Falcon3-1B1B
FALCON·32K CTX· CHAT
971
TOK/S · 0% VRAM
S
InternLM2 1B1B
INTERNLM·32K CTX· CHAT
971
TOK/S · 0% VRAM
S
InternVL3 1B1B
OTHER·32K CTX· CHAT· VISION
971
TOK/S · 0% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
882
TOK/S · 0% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
809
TOK/S · 0% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
809
TOK/S · 0% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
809
TOK/S · 0% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
809
TOK/S · 0% VRAM
S
DeepSeek Coder 1.3B1.3B
DEEPSEEK·16K CTX· CODING
747
TOK/S · 0% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
747
TOK/S · 0% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
747
TOK/S · 0% VRAM
S
Phi-1 1.3B1.3B
PHI·2K CTX· CODING
747
TOK/S · 0% VRAM
S
Phi-1.5 1.3B1.3B
PHI·2K CTX· CHAT· CODING
747
TOK/S · 0% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Apple M4 Ultra (384GB).

▸ DEVICE UNDER TEST

Apple M4 Ultra (384GB)384 GB VRAM.

APPLE M4 ULTRA (384GB) SPEC
BRAND
Apple
VRAM
384 GB LPDDR5X
BANDWIDTH
1092 GB/s
FP16 COMPUTE
40 TFLOPS
FP32 COMPUTE
40 TFLOPS
TDP
215 W
ARCHITECTURE
M4 Ultra
MSRP
$9999
▸ AI CAPABILITY
411/ 428 models @ Q4

With 384 GB VRAM and 1092 GB/s bandwidth, this GPU handles models up to 480.2B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~125 tok/s.

§ 01TOP MODELS FOR APPLE M4 ULTRA (384GB)
411 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Qwen3-Coder-480B-A35B480.2B294.0 GB2845.1
Qwen3-Coder 480B-A35B480B293.9 GB2845.1
Llama 3.1 405B405B248.0 GB248.4
Hermes 4 405B405B248.0 GB233.4
Cogito v2 405B405B248.0 GB2
Llama-4-Maverick-17B-128E400B245.0 GB5746.1
Jamba Large 1.6398B243.8 GB1011.0
Qwen3.5-397B-A17B397B243.2 GB5768.4
Hunyuan-Large389B238.3 GB1973.9
GLM 4.7358B219.3 GB3064.4
GLM 4.6357B218.7 GB8141.4
GLM 4.5355B217.5 GB8147.5
Nemotron-4 340B340B208.3 GB365.4
MiMo V2 Flash 309B309B189.4 GB6572.4
ERNIE-4.5-300B-A47B-Paddle300.5B184.2 GB337.4
ERNIE 4.5 300B-A47B300B183.9 GB2149.8
Hunyuan Hy3-preview295B180.8 GB4662.7
DeepSeek V4 Flash284B174.1 GB7559.3
DeepSeek V2 236B236B144.7 GB4640.4
DeepSeek-V2.5 236B236B144.7 GB4664.0