FitMyLLM
INTEL· XE-HPG

Intel Arc Pro A60

Running LLMs on the Arc Pro A60 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
12 GB
ENTRY-LEVEL
BANDWIDTH
384
GB/S
MODELS Q4
248/449
55%
7B Q4 SPEED
~18
GOOD
▸ MODEL COVERAGE @ Q455% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~41
TOK/S
7B
~18
TOK/S
14B
~9
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
VRAM
12 GB
BANDWIDTH
384 GB/s
FP16 COMPUTE
16.8 TFLOPS
TDP
130W
MEMORY
GDDR6
ARCHITECTURE
Xe-HPG
COMPUTE UNITS
128
PCIE
Gen 4 x16
101
FAST MODELS · >30 TOK/S
Real-time chat speed
228
USABLE · >10 TOK/S
Comfortable for all tasks
248
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Arc Pro A60 (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· 248
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
1365
TOK/S · 5% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
910
TOK/S · 5% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
878
TOK/S · 5% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
878
TOK/S · 5% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
480
TOK/S · 5% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
455
TOK/S · 5% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
403
TOK/S · 6% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
367
TOK/S · 6% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
367
TOK/S · 6% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
367
TOK/S · 6% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
351
TOK/S · 6% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
341
TOK/S · 6% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
323
TOK/S · 6% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
246
TOK/S · 7% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
246
TOK/S · 7% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
236
TOK/S · 7% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
216
TOK/S · 7% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
216
TOK/S · 7% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
205
TOK/S · 7% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
205
TOK/S · 7% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
205
TOK/S · 7% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
198
TOK/S · 7% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
166
TOK/S · 8% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
164
TOK/S · 8% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
152
TOK/S · 8% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
141
TOK/S · 9% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
131
TOK/S · 9% VRAM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
114
TOK/S · 10% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
112
TOK/S · 10% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
102
TOK/S · 10% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
102
TOK/S · 10% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
102
TOK/S · 10% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
102
TOK/S · 10% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
95
TOK/S · 11% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
95
TOK/S · 11% VRAM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
95
TOK/S · 11% VRAM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
91
TOK/S · 11% VRAM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
87
TOK/S · 11% VRAM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
87
TOK/S · 11% VRAM
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
82
TOK/S · 46% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Arc Pro A60.

WHAT THIS CARD IS WORTH

Arc Pro A60 holds 248 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is Ling-lite 16.8B at IQ4_XS.

TOKENS/SEC PER $100
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
what memory costs on this card
THE BIGGEST IT TAKES
Ling-lite 16.8B16.8B · IQ4_XS10.7 GB52 tok/sEST
DeepSeek V2 Lite 16B16B · Q4_K_S10.6 GB49 tok/sEST
StarCoder2 15B15.96B · Q4_K_S10.7 GB7 tok/sEST
DeepSeek-Coder-V2-Lite 15.7B15.7B · Q4_K_S10.4 GB49 tok/sEST
DeepSeek R1 Distill Qwen 14B14.8B · Q4_K_S10.4 GB8 tok/sEST
DeepCoder 14B14.8B · Q4_K_S10.4 GB8 tok/sEST
Qwen2.5-Coder-14B14.8B · Q4_K_S10.4 GB8 tok/sEST
Qwen2.5-14B14.8B · Q4_K_S10.4 GB8 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M4 Pro (24GB)16 GB$1,3992.5 tok/s per $100
M4 (24GB)16 GB$6992.9 tok/s per $100
M2 (24GB)16 GB$9991.7 tok/s per $100
M3 (24GB)16 GB$9991.7 tok/s per $100
Arc Pro A6012 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

Intel Arc Pro A60 — 12 GB VRAM.

ARC PRO A60 SPEC
BRAND
Intel
VRAM
12 GB GDDR6
BANDWIDTH
384 GB/s
FP16 COMPUTE
16.8 TFLOPS
FP32 COMPUTE
8.4 TFLOPS
TDP
130 W
ARCHITECTURE
Xe-HPG
▸ AI CAPABILITY
248/ 449 models @ Q4

With 12 GB VRAM and 384 GB/s bandwidth, this GPU handles models up to 14.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~44 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR ARC PRO A60
248 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB843.9
DeepCoder 14B14.8B9.5 GB838.7
Qwen2.5-Coder-14B14.8B9.5 GB841.3
Qwen2.5-14B14.8B9.5 GB841.3
Qwen3 14B14.8B9.5 GB845.7
phi-4 14B14.66B9.4 GB833.7
Phi-4-reasoning 14B14.66B9.4 GB833.7
Phi-4-reasoning-plus 14B14.66B9.4 GB875.5
Phi-3-medium-14b14B9.0 GB933.7
Qwen 1.5 14B14B9.0 GB941.3
Ministral 3 14B Reasoning13.95B9.0 GB9
Baichuan2 13B13B8.4 GB923.6
Llama 2 13B13B8.4 GB917.2
CodeLlama 13B13B8.4 GB919.7
Vicuna 13B13B8.4 GB911.8
LLaMA 1 13B13B8.4 GB932.9
OPT 13B13B8.4 GB935.8
Orca 2 13B13B8.4 GB925.4
WizardCoder Python 13B13B8.4 GB960.1
WizardLM 13B13B8.4 GB919.5