FitMyLLM
APPLE· M5 MAX

Apple M5 Max (128GB)

Running LLMs on the Apple M5 Max (128GB) — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

USABLE MEMORY
96 GB
FLAGSHIP
BANDWIDTH
614
GB/S
MODELS Q4
392/449
87%
7B Q4 SPEED
~78
BLAZING
▸ MODEL COVERAGE @ Q487% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~182
TOK/S
7B
~78
TOK/S
14B
~39
TOK/S
32B
~17
TOK/S
70B
~8
TOK/S
▸ MEASURED RIG REPORTS

We rent the machine and time every model on it: decode, VRAM peak, concurrency, watts and cost per million tokens. Yours may already be one of them — and two are free to read in full.

SEE THE REPORTS
▸ SPECIFICATIONS
USABLE MEMORY
96 GB of 128
BANDWIDTH
614 GB/s
FP16 COMPUTE
18 TFLOPS
TDP
45W
MEMORY
Unified
ARCHITECTURE
M5 Max
296
FAST MODELS · >30 TOK/S
Real-time chat speed
357
USABLE · >10 TOK/S
Comfortable for all tasks
392
TOTAL COMPATIBLE
Fit in usable memory at Q4
▸ DON’T WANT TO BUY?

Test M5 Max (128GB) (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· 392
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
6064
TOK/S · 1% MEM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
4043
TOK/S · 1% MEM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
3898
TOK/S · 1% MEM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
3898
TOK/S · 1% MEM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
2132
TOK/S · 1% MEM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
2021
TOK/S · 1% MEM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
1789
TOK/S · 1% MEM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
1629
TOK/S · 1% MEM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
1629
TOK/S · 1% MEM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
1629
TOK/S · 1% MEM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
1559
TOK/S · 1% MEM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
1516
TOK/S · 1% MEM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
1436
TOK/S · 1% MEM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
1092
TOK/S · 1% MEM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
1092
TOK/S · 1% MEM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
1050
TOK/S · 1% MEM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
961
TOK/S · 1% MEM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
961
TOK/S · 1% MEM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
910
TOK/S · 1% MEM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
910
TOK/S · 1% MEM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
910
TOK/S · 1% MEM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
880
TOK/S · 1% MEM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
738
TOK/S · 1% MEM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
728
TOK/S · 1% MEM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
674
TOK/S · 1% MEM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
627
TOK/S · 1% MEM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
581
TOK/S · 1% MEM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
505
TOK/S · 1% MEM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
496
TOK/S · 1% MEM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
455
TOK/S · 1% MEM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
455
TOK/S · 1% MEM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
455
TOK/S · 1% MEM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
455
TOK/S · 1% MEM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
420
TOK/S · 1% MEM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
420
TOK/S · 1% MEM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
420
TOK/S · 1% MEM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
404
TOK/S · 1% MEM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
384
TOK/S · 1% MEM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
384
TOK/S · 1% MEM
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
364
TOK/S · 6% MEM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Apple M5 Max (128GB).

WHAT THIS CARD IS WORTH

M5 Max (128GB) holds 392 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is dots.llm1.inst 142.8B at IQ4_XS.

TOKENS/SEC PER $100
8B at Q4_K_M, so cards compare like for like
USABLE MEMORY PER $100
what memory costs on this card
THE BIGGEST IT TAKES
dots.llm1.inst 142.8B142.8B · IQ4_XS86.0 GB3 tok/sEST
WizardLM 2 8x22B141B · Q4_K_S85.7 GB12 tok/sEST
Mixtral-8x22B140.6B · Q4_K_S85.5 GB12 tok/sEST
DBRX 132B132B · Q4_K_M83.8 GB12 tok/sEST
Mistral Medium 3.5128B · Q4_K_M82.1 GB3 tok/sEST
Pixtral Large 124B124B · Q4_K_M79.7 GB4 tok/sEST
Nemotron 3 Super 120B-A12B123.61B · Q4_K_M78.4 GB37 tok/sEST
Mistral-Large 123B123B · Q4_K_M79.1 GB4 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
Radeon Instinct MI300128 GB$12,0005.8 tok/s per $100
Radeon Instinct MI250128 GB$12,0002.9 tok/s per $100
Instinct MI250X128 GB$10,0003.5 tok/s per $100
M1 Ultra (128GB)96 GB$4,9991.4 tok/s per $100
M5 Max (128GB)96 GB

Neighbours in memory rather than in price: memory decides whether a card can do the job at all, so two cards of the same size at different prices is the comparison you are making. Ordered by memory, then bandwidth — not by the value column, which is worked out from bandwidth and price alone and therefore rewards a cheap card whatever its software stack does to that bandwidth in practice. Read it as one input, not as a ranking.

▸ DEVICE UNDER TEST

Apple M5 Max (128GB) — 128 GB unified, 96 GB usable.

APPLE M5 MAX (128GB) SPEC
BRAND
Apple
UNIFIED MEMORY
128 GB
USABLE BY A MODEL
96 GB
BANDWIDTH
614 GB/s
FP16 COMPUTE
18 TFLOPS
TDP
45 W
ARCHITECTURE
M5 Max
▸ AI CAPABILITY
392/ 449 models @ Q4

With 96 GB of its 128 GB reaching a model and 614 GB/s bandwidth, this machine handles models up to 132B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~70 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR APPLE M5 MAX (128GB)
392 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
DBRX 132B132B81.2 GB1546.3
Mistral Medium 3.5128B78.7 GB446.6
Pixtral Large 124B124B76.3 GB439.3
Nemotron 3 Super 120B-A12B123.61B76.0 GB4553.2
Mistral-Large 123B123B75.7 GB433.5
Devstral 2 123B123B75.7 GB438.1
Qwen 3.5 122B A10B122B75.1 GB5556.8
Nemotron 3 Super 120B120B73.8 GB4557.3
Mistral Small 4 119B119B73.2 GB8450.2
GPT-OSS 120B117B72.0 GB10754.1
Command A 111B111B68.3 GB527.6
GLM 4.5 Air110B67.7 GB4551.0
Qwen 1.5 110B110B67.7 GB533.4
Llama 4 Scout 17B-16E109B67.1 GB3233.9
Cogito v2 109B MoE109B67.1 GB32
Ling 2.6 Flash107.49B66.2 GB7436.8
Sarvam 105B105B64.7 GB548.0
Command-R+ 104B104B64.1 GB552.7
Llama-3.2-90B-Vision-Instruct90B55.5 GB648.5
Hunyuan A13B80B49.4 GB4281.1