FitMyLLM
▸ APPLE· M4 PRO

Apple M4 Pro (24GB)

Running LLMs on the Apple M4 Pro (24GB) — the long read: which models fit at which quantisation, and the settings worth changing. · Or what else $1,399 buys

USABLE MEMORY
16 GB
MID-RANGE
BANDWIDTH
273
GB/S
MODELS Q4
262/449
58%
7B Q4 SPEED
~35
FAST
▸ MODEL COVERAGE @ Q458% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~81
TOK/S
7B
~35
TOK/S
14B
~17
TOK/S
32B
—
18.0GB NEEDED
70B
—
39.4GB NEEDED
▸ 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
16 GB of 24
BANDWIDTH
273 GB/s
FP16 COMPUTE
9.2 TFLOPS
TDP
30W
MEMORY
Unified
ARCHITECTURE
M4 Pro
MSRP
$1,399
198
FAST MODELS · >30 TOK/S
Real-time chat speed
262
USABLE · >10 TOK/S
Comfortable for all tasks
262
TOTAL COMPATIBLE
Fit in usable memory at Q4
▸ DON’T WANT TO BUY?

Test M4 Pro (24GB) (or anything bigger) without committing. Pay by the second, cancel anytime.

Rent on Vast.ai →Or RunPod →

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· 262
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
2696
TOK/S · 3% MEM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1798
TOK/S · 4% MEM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
1733
TOK/S · 4% MEM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
1733
TOK/S · 4% MEM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
948
TOK/S · 4% MEM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
899
TOK/S · 4% MEM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
796
TOK/S · 4% MEM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
724
TOK/S · 4% MEM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
724
TOK/S · 4% MEM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
724
TOK/S · 4% MEM
›
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
693
TOK/S · 4% MEM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
674
TOK/S · 4% MEM
›
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
639
TOK/S · 5% MEM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
485
TOK/S · 5% MEM
›
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
485
TOK/S · 5% MEM
›
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
467
TOK/S · 5% MEM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
427
TOK/S · 5% MEM
›
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
427
TOK/S · 5% MEM
›
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
404
TOK/S · 5% MEM
›
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
404
TOK/S · 5% MEM
›
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
404
TOK/S · 5% MEM
›
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
391
TOK/S · 5% MEM
›
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
328
TOK/S · 6% MEM
›
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
324
TOK/S · 6% MEM
›
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
300
TOK/S · 6% MEM
›
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
279
TOK/S · 6% MEM
›
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
258
TOK/S · 7% MEM
›
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
225
TOK/S · 7% MEM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
221
TOK/S · 7% MEM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
202
TOK/S · 8% MEM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
202
TOK/S · 8% MEM
›
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
202
TOK/S · 8% MEM
›
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
202
TOK/S · 8% MEM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
187
TOK/S · 8% MEM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
187
TOK/S · 8% MEM
›
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
187
TOK/S · 8% MEM
›
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
180
TOK/S · 8% MEM
›
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
171
TOK/S · 8% MEM
›
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
171
TOK/S · 8% MEM
›
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
162
TOK/S · 35% MEM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Apple M4 Pro (24GB).

▸Find models for Apple M4 Pro (24GB)Compare with other GPUs
WHAT THIS CARD IS WORTH

M4 Pro (24GB) holds 262 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is ERNIE 4.5 21B A3B at IQ4_XS.

TOKENS/SEC PER $100
2.5
8B at Q4_K_M, so cards compare like for like
USABLE MEMORY PER $100
1.14 GB
what memory costs on this card
THE BIGGEST IT TAKES
ERNIE 4.5 21B A3B21.95B · IQ4_XS13.9 GB94 tok/sEST
GPT-OSS 20B21B · Q4_K_S13.8 GB75 tok/sEST
Reka Flash 321B · Q4_K_S14.2 GB13 tok/sEST
Reka Flash 3.121B · Q4_K_S14.2 GB13 tok/sEST
InternLM2 20B19.8B · Q4_K_M14.3 GB13 tok/sEST
InternLM2.5 20B19.8B · Q4_K_M14.3 GB13 tok/sEST
Ling-lite 16.8B16.8B · Q5_K_M13.6 GB92 tok/sEST
DeepSeek V2 Lite 16B16B · Q5_K_M12.9 GB92 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M1 Max (32GB)21 GB$1,4993.0 tok/s per $100
M2 Max (32GB)21 GB$1,7992.5 tok/s per $100
M2 Pro (32GB)21 GB$1,4991.9 tok/s per $100
M4 (32GB)21 GB$1,1991.7 tok/s per $100
M4 Pro (24GB)16 GB$1,3992.5 tok/s per $100

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 M4 Pro (24GB) — 24 GB unified, 16 GB usable.

▸ APPLE M4 PRO (24GB) SPEC
BRAND
Apple
UNIFIED MEMORY
24 GB
USABLE BY A MODEL
16 GB
BANDWIDTH
273 GB/s
FP16 COMPUTE
9.2 TFLOPS
FP32 COMPUTE
9.2 TFLOPS
TDP
30 W
ARCHITECTURE
M4 Pro
MSRP
$1399
▸ AI CAPABILITY
262/ 449 models @ Q4

With 16 GB of its 24 GB reaching a model and 273 GB/s bandwidth, this machine handles models up to 19.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~31 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR APPLE M4 PRO (24GB)
262 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
InternLM2 20B19.8B12.6 GB1245.1
InternLM2.5 20B19.8B12.6 GB1250.9
Ling-lite 16.8B16.8B10.8 GB101—
DeepSeek V2 Lite 16B16B10.3 GB10138.0
StarCoder2 15B15.96B10.2 GB1526.5
DeepSeek-Coder-V2-Lite 15.7B15.7B10.1 GB10143.0
DeepSeek-VL2 Small 16B15.7B10.1 GB10143.1
StarCoder 15B15.5B10.0 GB1621.0
InternVL3 14B15.12B9.7 GB1638.1
Phi-4-reasoning-vision 15B15B9.7 GB1642.8
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB1643.9
DeepCoder 14B14.8B9.5 GB1638.7
Qwen2.5-Coder-14B14.8B9.5 GB1641.3
Qwen2.5-14B14.8B9.5 GB1641.3
Qwen3 14B14.8B9.5 GB1645.7
phi-4 14B14.66B9.4 GB1733.7
Phi-4-reasoning 14B14.66B9.4 GB1733.7
Phi-4-reasoning-plus 14B14.66B9.4 GB1775.5
Ministral 3 14B14B9.0 GB1725.9
Phi-3-medium-14b14B9.0 GB1733.7