FitMyLLM
▸ AMD· RDNA 1.0

AMD Radeon Pro 5300M

Running LLMs on the Radeon Pro 5300M — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
4 GB
BUDGET
BANDWIDTH
192
GB/S
MODELS Q4
97/449
22%
7B Q4 SPEED
~24
GOOD
▸ MODEL COVERAGE @ Q422% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~57
TOK/S
7B
—
3.9GB NEEDED
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
4 GB
BANDWIDTH
192 GB/s
FP16 COMPUTE
6.4 TFLOPS
TDP
85W
MEMORY
GDDR6
ARCHITECTURE
RDNA 1.0
STREAM PROCESSORS
1,280
COMPUTE UNITS
20
PCIE
Gen 4 x8
97
FAST MODELS · >30 TOK/S
Real-time chat speed
97
USABLE · >10 TOK/S
Comfortable for all tasks
97
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Radeon Pro 5300M (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· 97
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
1896
TOK/S · 14% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1264
TOK/S · 14% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
1219
TOK/S · 14% VRAM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
1219
TOK/S · 14% VRAM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
667
TOK/S · 16% VRAM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
632
TOK/S · 16% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
560
TOK/S · 17% VRAM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
509
TOK/S · 17% VRAM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
509
TOK/S · 17% VRAM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
509
TOK/S · 17% VRAM
›
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
488
TOK/S · 18% VRAM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
474
TOK/S · 18% VRAM
›
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
449
TOK/S · 18% VRAM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
341
TOK/S · 20% VRAM
›
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
341
TOK/S · 20% VRAM
›
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
328
TOK/S · 20% VRAM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
300
TOK/S · 21% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
300
TOK/S · 21% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
284
TOK/S · 21% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
284
TOK/S · 21% VRAM
›
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
284
TOK/S · 21% VRAM
›
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
275
TOK/S · 22% VRAM
›
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
231
TOK/S · 24% VRAM
›
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
228
TOK/S · 24% VRAM
›
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
211
TOK/S · 25% VRAM
›
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
196
TOK/S · 26% VRAM
›
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
182
TOK/S · 27% VRAM
›
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
158
TOK/S · 29% VRAM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
155
TOK/S · 29% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
142
TOK/S · 31% VRAM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
142
TOK/S · 31% VRAM
›
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
142
TOK/S · 31% VRAM
›
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
142
TOK/S · 31% VRAM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
131
TOK/S · 32% VRAM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
131
TOK/S · 32% VRAM
›
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
131
TOK/S · 32% VRAM
›
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
126
TOK/S · 33% VRAM
›
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
120
TOK/S · 34% VRAM
›
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
120
TOK/S · 34% VRAM
›
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
114
TOK/S · 35% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Radeon Pro 5300M.

WHAT THIS CARD IS WORTH

Radeon Pro 5300M holds 97 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is InternLM2 5B at Q4_K_M.

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
InternLM2 5B4.5B · Q4_K_M3.6 GB31 tok/sEST
Ministral 3 3B Reasoning4.25B · Q4_K_M3.3 GB32 tok/sEST
Qwen3 4B4B · Q4_K_M3.3 GB34 tok/sEST
Qwen3-4B Instruct 25074B · Q4_K_M3.3 GB34 tok/sEST
Qwen3-Embedding 4B4B · Q4_K_M3.3 GB34 tok/sEST
Nemotron 3 Nano 4B3.97B · Q4_K_M3.4 GB34 tok/sEST
Ministral 3 3B3.85B · Q5_K_M3.5 GB31 tok/sEST
phi-3-mini-4k 3.8B3.8B · Q4_K_M3.4 GB35 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
RTX 3070 Ti8 GB$49912.6 tok/s per $100
RTX 3060 Ti GDDR6X8 GB$39915.8 tok/s per $100
RTX 3070 Ti 8 GB GA1028 GB$59910.5 tok/s per $100
Arc A7508 GB$19910.1 tok/s per $100
Radeon Pro 5300M4 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 Radeon Pro 5300M — 4 GB VRAM.

▸ RADEON PRO 5300M SPEC
BRAND
AMD
VRAM
4 GB GDDR6
BANDWIDTH
192 GB/s
FP16 COMPUTE
6.4 TFLOPS
FP32 COMPUTE
3.2 TFLOPS
STREAM PROCESSORS
1,280
TDP
85 W
ARCHITECTURE
RDNA 1.0
▸ AI CAPABILITY
97/ 449 models @ Q4

With 4 GB VRAM and 192 GB/s bandwidth, this GPU handles models up to 4.25B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~22 tok/s.

§ 01TOP MODELS FOR RADEON PRO 5300M
97 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Ministral 3 3B Reasoning4.25B3.1 GB40—
Qwen 1.5 4B4B2.9 GB4312.6
Qwen3 4B4B2.9 GB4340.7
Qwen3-4B Instruct 25074B2.9 GB4337.2
Qwen3-Embedding 4B4B2.9 GB43—
Nemotron 3 Nano 4B3.97B2.9 GB4332.0
Ministral 3 3B3.85B2.8 GB4421.4
Phi-3.5 Mini 3.8B3.82B2.8 GB4546.6
phi-3-mini-4k 3.8B3.8B2.8 GB4530.5
Phi-4-mini 3.8B3.8B2.8 GB4549.0
Cogito 3B3.61B2.7 GB4722.1
Falcon3-3B3.23B2.5 GB5325.7
granite-4.0-h-micro 3.2B3.2B2.4 GB5318.4
Llama-3.2-3B3.2B2.4 GB5317.9
Falcon-H1 3B3.15B2.4 GB5449.5
Qwen 2.5 3B3.1B2.4 GB5537.2
SmolLM3-3B3.1B2.4 GB5530.5
Ministral 3B3B2.3 GB5729.6
StarCoder2 3B3B2.3 GB579.5
Granite 4.1 3B3B2.3 GB5716.6