FitMyLLM
QUALCOMM· ADRENO 600

Qualcomm Adreno 640

Running LLMs on the Qualcomm Adreno 640 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
8 GB
ENTRY-LEVEL
BANDWIDTH
34.1
GB/S
MODELS Q4
201/449
45%
7B Q4 SPEED
~4
SLOW
▸ MODEL COVERAGE @ Q445% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~10
TOK/S
7B
~4
TOK/S
14B
7.9GB NEEDED
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
8 GB
BANDWIDTH
34.1 GB/s
FP16 COMPUTE
1 TFLOPS
TDP
7W
MEMORY
Shared
ARCHITECTURE
Adreno 600
27
FAST MODELS · >30 TOK/S
Real-time chat speed
85
USABLE · >10 TOK/S
Comfortable for all tasks
201
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Qualcomm Adreno 640 (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· 201
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
337
TOK/S · 7% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
225
TOK/S · 7% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
217
TOK/S · 7% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
217
TOK/S · 7% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
118
TOK/S · 8% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
112
TOK/S · 8% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
99
TOK/S · 8% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
90
TOK/S · 9% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
90
TOK/S · 9% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
90
TOK/S · 9% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
87
TOK/S · 9% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
84
TOK/S · 9% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
80
TOK/S · 9% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
61
TOK/S · 10% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
61
TOK/S · 10% VRAM
A
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
58
TOK/S · 10% VRAM
A
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
53
TOK/S · 10% VRAM
A
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
53
TOK/S · 10% VRAM
A
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
51
TOK/S · 11% VRAM
A
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
51
TOK/S · 11% VRAM
A
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
51
TOK/S · 11% VRAM
A
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
49
TOK/S · 11% VRAM
A
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
41
TOK/S · 12% VRAM
A
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
40
TOK/S · 12% VRAM
B
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
37
TOK/S · 12% VRAM
B
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
35
TOK/S · 13% VRAM
B
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
32
TOK/S · 13% VRAM
B
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
28
TOK/S · 15% VRAM
B
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
28
TOK/S · 14% VRAM
B
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
25
TOK/S · 15% VRAM
B
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
25
TOK/S · 15% VRAM
B
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
25
TOK/S · 15% VRAM
B
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
25
TOK/S · 15% VRAM
C
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
23
TOK/S · 16% VRAM
C
OPT 1.3B1.3B
OPT·2K CTX· CHAT
23
TOK/S · 16% VRAM
C
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
23
TOK/S · 16% VRAM
C
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
22
TOK/S · 16% VRAM
C
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
21
TOK/S · 17% VRAM
C
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
21
TOK/S · 17% VRAM
C
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
20
TOK/S · 70% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Qualcomm Adreno 640.

WHAT THIS CARD IS WORTH

Qualcomm Adreno 640 holds 201 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is Falcon3-10B 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
Falcon3-10B10.3B · IQ4_XS7.1 GB4 tok/sEST
Bamba 9B v29.78B · Q4_K_M7.2 GB4 tok/sEST
RecurrentGemma 9B9.63B · Q4_K_M6.8 GB4 tok/sEST
glm-4-9b9.4B · Q4_K_M6.6 GB4 tok/sEST
Yi 1.5 9B9B · Q4_K_M6.5 GB5 tok/sEST
Yi Coder 9B9B · Q4_K_M6.5 GB5 tok/sEST
Ministral 3 8B8.92B · Q4_K_M6.6 GB5 tok/sEST
Ministral 3 8B Reasoning8.92B · Q4_K_M6.6 GB5 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M3 Pro (18GB)12 GB$1,5991.5 tok/s per $100
M1 Pro (16GB)11 GB$9992.9 tok/s per $100
M2 Pro (16GB)11 GB$1,2992.2 tok/s per $100
M4 (16GB)11 GB$4994.0 tok/s per $100
Qualcomm Adreno 6408 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

Qualcomm Adreno 640 — 8 GB VRAM.

QUALCOMM ADRENO 640 SPEC
BRAND
Apple
VRAM
8 GB Shared
BANDWIDTH
34.1 GB/s
FP16 COMPUTE
1 TFLOPS
TDP
7 W
ARCHITECTURE
Adreno 600
▸ AI CAPABILITY
201/ 449 models @ Q4

With 8 GB VRAM and 34.1 GB/s bandwidth, this GPU handles models up to 9.63B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~4 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR QUALCOMM ADRENO 640
201 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
RecurrentGemma 9B9.63B6.4 GB335.0
glm-4-9b9.4B6.2 GB320.5
gemma-2-9b9.2B6.1 GB330.2
Yi 1.5 9B9B6.0 GB330.3
Yi Coder 9B9B6.0 GB335.8
Ministral 3 8B8.92B5.9 GB325.7
Ministral 3 8B Reasoning8.92B5.9 GB3
NVIDIA-Nemotron-Nano-9B-v28.9B5.9 GB344.2
InternLM3 8B Instruct8.8B5.9 GB338.7
Gemma 1 7B8.54B5.7 GB424.7
CodeGemma 7B8.54B5.7 GB440.2
LFM2 8B A1B8.3B5.6 GB2024.3
Seed-Coder 8B Instruct8.25B5.5 GB434.1
Seed-Coder 8B Reasoning8.25B5.5 GB432.9
DeepSeek R1-0528 Qwen3 8B8.2B5.5 GB436.3
Qwen3-8B8.2B5.5 GB443.3
Granite 3.0 8B8.17B5.5 GB436.4
Granite 3.1 8B8.17B5.5 GB438.6
Command-R7B8.03B5.4 GB435.3
Aya Expanse 8B8B5.4 GB427.8