FitMyLLM
AMD· RDNA 2.0

AMD Steam Deck GPU

VRAM
16 GB
MID-RANGE
BANDWIDTH
176
GB/S
MODELS Q4
262/449
58%
7B Q4 SPEED
~22
GOOD
▸ 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
~52
TOK/S
7B
~22
TOK/S
14B
~11
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
16 GB
BANDWIDTH
176 GB/s
FP16 COMPUTE
3.3 TFLOPS
TDP
15W
MEMORY
LPDDR5
ARCHITECTURE
RDNA 2.0
STREAM PROCESSORS
512
COMPUTE UNITS
8
113
FAST MODELS · >30 TOK/S
Real-time chat speed
260
USABLE · >10 TOK/S
Comfortable for all tasks
262
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Steam Deck GPU (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· 262
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
1738
TOK/S · 3% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1159
TOK/S · 4% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
1117
TOK/S · 4% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
1117
TOK/S · 4% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
611
TOK/S · 4% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
579
TOK/S · 4% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
513
TOK/S · 4% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
467
TOK/S · 4% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
467
TOK/S · 4% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
467
TOK/S · 4% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
447
TOK/S · 4% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
435
TOK/S · 4% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
412
TOK/S · 5% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
313
TOK/S · 5% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
313
TOK/S · 5% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
301
TOK/S · 5% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
275
TOK/S · 5% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
275
TOK/S · 5% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
261
TOK/S · 5% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
261
TOK/S · 5% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
261
TOK/S · 5% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
252
TOK/S · 5% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
211
TOK/S · 6% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
209
TOK/S · 6% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
193
TOK/S · 6% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
180
TOK/S · 6% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
166
TOK/S · 7% VRAM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
145
TOK/S · 7% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
142
TOK/S · 7% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
130
TOK/S · 8% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
130
TOK/S · 8% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
130
TOK/S · 8% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
130
TOK/S · 8% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
120
TOK/S · 8% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
120
TOK/S · 8% VRAM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
120
TOK/S · 8% VRAM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
116
TOK/S · 8% VRAM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
110
TOK/S · 8% VRAM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
110
TOK/S · 8% VRAM
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
104
TOK/S · 35% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Steam Deck GPU.

WHAT THIS CARD IS WORTH

Steam Deck GPU 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
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
what memory costs on this card
THE BIGGEST IT TAKES
ERNIE 4.5 21B A3B21.95B · IQ4_XS13.9 GB41 tok/sEST
GPT-OSS 20B21B · Q4_K_S13.8 GB35 tok/sEST
Reka Flash 321B · Q4_K_S14.2 GB9 tok/sEST
Reka Flash 3.121B · Q4_K_S14.2 GB9 tok/sEST
InternLM2 20B19.8B · Q4_K_M14.3 GB9 tok/sEST
InternLM2.5 20B19.8B · Q4_K_M14.3 GB9 tok/sEST
Ling-lite 16.8B16.8B · Q5_K_M13.6 GB41 tok/sEST
DeepSeek V2 Lite 16B16B · Q5_K_M12.9 GB41 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
Steam Deck GPU16 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

AMD Steam Deck GPU — 16 GB VRAM.

STEAM DECK GPU SPEC
BRAND
AMD
VRAM
16 GB LPDDR5
BANDWIDTH
176 GB/s
FP16 COMPUTE
3.3 TFLOPS
FP32 COMPUTE
1.6 TFLOPS
STREAM PROCESSORS
512
TDP
15 W
ARCHITECTURE
RDNA 2.0
▸ AI CAPABILITY
262/ 449 models @ Q4

With 16 GB VRAM and 176 GB/s bandwidth, this GPU handles models up to 19.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~20 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR STEAM DECK GPU
262 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
InternLM2 20B19.8B12.6 GB845.1
InternLM2.5 20B19.8B12.6 GB850.9
Ling-lite 16.8B16.8B10.8 GB65
DeepSeek V2 Lite 16B16B10.3 GB6538.0
StarCoder2 15B15.96B10.2 GB1026.5
DeepSeek-Coder-V2-Lite 15.7B15.7B10.1 GB6543.0
DeepSeek-VL2 Small 16B15.7B10.1 GB6543.1
StarCoder 15B15.5B10.0 GB1021.0
InternVL3 14B15.12B9.7 GB1038.1
Phi-4-reasoning-vision 15B15B9.7 GB1042.8
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB1143.9
DeepCoder 14B14.8B9.5 GB1138.7
Qwen2.5-Coder-14B14.8B9.5 GB1141.3
Qwen2.5-14B14.8B9.5 GB1141.3
Qwen3 14B14.8B9.5 GB1145.7
phi-4 14B14.66B9.4 GB1133.7
Phi-4-reasoning 14B14.66B9.4 GB1133.7
Phi-4-reasoning-plus 14B14.66B9.4 GB1175.5
Ministral 3 14B14B9.0 GB1125.9
Phi-3-medium-14b14B9.0 GB1133.7