FitMyLLM
▸ AMD· RDNA 1.0

AMD Radeon RX 5300 OEM

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

VRAM
3 GB
BUDGET
BANDWIDTH
168
GB/S
MODELS Q4
81/449
18%
7B Q4 SPEED
~21
GOOD
▸ MODEL COVERAGE @ Q418% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~50
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
3 GB
BANDWIDTH
168 GB/s
FP16 COMPUTE
9.3 TFLOPS
TDP
100W
MEMORY
GDDR6
ARCHITECTURE
RDNA 1.0
STREAM PROCESSORS
1,408
COMPUTE UNITS
22
PCIE
Gen 4 x8
81
FAST MODELS · >30 TOK/S
Real-time chat speed
81
USABLE · >10 TOK/S
Comfortable for all tasks
81
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Radeon RX 5300 OEM (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· 81
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
1659
TOK/S · 18% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1106
TOK/S · 19% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
1067
TOK/S · 19% VRAM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
1067
TOK/S · 19% VRAM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
583
TOK/S · 21% VRAM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
553
TOK/S · 22% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
490
TOK/S · 22% VRAM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
446
TOK/S · 23% VRAM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
446
TOK/S · 23% VRAM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
446
TOK/S · 23% VRAM
›
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
427
TOK/S · 23% VRAM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
415
TOK/S · 24% VRAM
›
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
393
TOK/S · 24% VRAM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
299
TOK/S · 26% VRAM
›
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
299
TOK/S · 26% VRAM
›
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
287
TOK/S · 27% VRAM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
263
TOK/S · 28% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
263
TOK/S · 28% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
249
TOK/S · 29% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
249
TOK/S · 29% VRAM
›
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
249
TOK/S · 29% VRAM
›
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
241
TOK/S · 29% VRAM
›
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
202
TOK/S · 31% VRAM
›
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
199
TOK/S · 32% VRAM
›
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
184
TOK/S · 33% VRAM
›
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
172
TOK/S · 34% VRAM
›
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
159
TOK/S · 35% VRAM
›
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
138
TOK/S · 38% VRAM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
136
TOK/S · 39% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
124
TOK/S · 41% VRAM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
124
TOK/S · 41% VRAM
›
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
124
TOK/S · 41% VRAM
›
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
124
TOK/S · 41% VRAM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
115
TOK/S · 43% VRAM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
115
TOK/S · 43% VRAM
›
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
115
TOK/S · 43% VRAM
›
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
111
TOK/S · 44% VRAM
›
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
105
TOK/S · 45% VRAM
›
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
105
TOK/S · 45% VRAM
›
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
100
TOK/S · 47% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

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

WHAT THIS CARD IS WORTH

Radeon RX 5300 OEM holds 81 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is Falcon3-3B 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
Falcon3-3B3.23B · Q4_K_M2.6 GB35 tok/sEST
granite-4.0-h-micro 3.2B3.2B · Q4_K_M2.6 GB35 tok/sEST
Llama-3.2-3B3.2B · Q4_K_M2.7 GB35 tok/sEST
Falcon-H1 3B3.15B · Q5_K_M2.7 GB32 tok/sEST
Qwen 2.5 3B3.1B · Q5_K_M2.6 GB32 tok/sEST
SmolLM3-3B3.1B · Q4_K_M2.5 GB36 tok/sEST
Ministral 3B3B · Q4_K_M2.6 GB37 tok/sEST
StarCoder2 3B3B · Q5_K_M2.5 GB33 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M2 (8GB)5 GB$5992.8 tok/s per $100
M3 (8GB)5 GB$5992.8 tok/s per $100
M1 (8GB)5 GB$4992.2 tok/s per $100
RTX 2060 6GB6 GB$15026.7 tok/s per $100
Radeon RX 5300 OEM3 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 RX 5300 OEM — 3 GB VRAM.

▸ RADEON RX 5300 OEM SPEC
BRAND
AMD
VRAM
3 GB GDDR6
BANDWIDTH
168 GB/s
FP16 COMPUTE
9.3 TFLOPS
FP32 COMPUTE
4.6 TFLOPS
STREAM PROCESSORS
1,408
TDP
100 W
ARCHITECTURE
RDNA 1.0
▸ AI CAPABILITY
81/ 449 models @ Q4

With 3 GB VRAM and 168 GB/s bandwidth, this GPU handles models up to 3.1B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~19 tok/s.

§ 01TOP MODELS FOR RADEON RX 5300 OEM
81 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Qwen 2.5 3B3.1B2.4 GB4837.2
SmolLM3-3B3.1B2.4 GB4830.5
Ministral 3B3B2.3 GB5029.6
StarCoder2 3B3B2.3 GB509.5
Granite 4.1 3B3B2.3 GB5016.6
xLAM-2 3B Function-Calling3B2.3 GB50—
Jamba 2 3B3B2.3 GB50—
Dolly v2 3B2.8B2.2 GB535.6
StableLM Zephyr 3B2.79B2.2 GB5414.9
Zephyr 3B2.79B2.2 GB5414.4
OPT 2.7B2.7B2.1 GB5528.0
Phi-2 2.7B2.7B2.1 GB5524.1
Zamba2 2.7B2.7B2.1 GB5548.0
Granite 3.0 2B2.63B2.1 GB5735.8
gemma-2-2b2.6B2.1 GB5722.9
LFM2 2.6B2.6B2.1 GB5716.3
Granite 3.1 2B2.53B2.0 GB5937.8
Granite 3.3 2B2.53B2.0 GB5920.5
Gemma 1 2B2.51B2.0 GB5920.2
CodeGemma 2B2.51B2.0 GB5922.9