FitMyLLM
▸ AMD· GCN 2.0

AMD Radeon R9 290X2

Running LLMs on the Radeon R9 290X2 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
4 GB
BUDGET
BANDWIDTH
346
GB/S
MODELS Q4
97/449
22%
7B Q4 SPEED
~44
FAST
▸ 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
~103
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
346 GB/s
FP16 COMPUTE
5.6 TFLOPS
TDP
580W
MEMORY
GDDR5
ARCHITECTURE
GCN 2.0
STREAM PROCESSORS
2,816
COMPUTE UNITS
44
PCIE
Gen 3 x16
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 R9 290X2 (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
3417
TOK/S · 14% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
2278
TOK/S · 14% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
2197
TOK/S · 14% VRAM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
2197
TOK/S · 14% VRAM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
1201
TOK/S · 16% VRAM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
1139
TOK/S · 16% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
1008
TOK/S · 17% VRAM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
918
TOK/S · 17% VRAM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
918
TOK/S · 17% VRAM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
918
TOK/S · 17% VRAM
›
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
879
TOK/S · 18% VRAM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
854
TOK/S · 18% VRAM
›
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
809
TOK/S · 18% VRAM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
615
TOK/S · 20% VRAM
›
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
615
TOK/S · 20% VRAM
›
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
591
TOK/S · 20% VRAM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
541
TOK/S · 21% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
541
TOK/S · 21% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
513
TOK/S · 21% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
513
TOK/S · 21% VRAM
›
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
513
TOK/S · 21% VRAM
›
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
496
TOK/S · 22% VRAM
›
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
416
TOK/S · 24% VRAM
›
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
410
TOK/S · 24% VRAM
›
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
380
TOK/S · 25% VRAM
›
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
354
TOK/S · 26% VRAM
›
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
327
TOK/S · 27% VRAM
›
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
285
TOK/S · 29% VRAM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
280
TOK/S · 29% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
256
TOK/S · 31% VRAM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
256
TOK/S · 31% VRAM
›
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
256
TOK/S · 31% VRAM
›
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
256
TOK/S · 31% VRAM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
237
TOK/S · 32% VRAM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
237
TOK/S · 32% VRAM
›
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
237
TOK/S · 32% VRAM
›
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
228
TOK/S · 33% VRAM
›
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
217
TOK/S · 34% VRAM
›
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
217
TOK/S · 34% VRAM
›
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
205
TOK/S · 35% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Radeon R9 290X2.

WHAT THIS CARD IS WORTH

Radeon R9 290X2 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 GB56 tok/sEST
Ministral 3 3B Reasoning4.25B · Q4_K_M3.3 GB59 tok/sEST
Qwen3 4B4B · Q4_K_M3.3 GB61 tok/sEST
Qwen3-4B Instruct 25074B · Q4_K_M3.3 GB61 tok/sEST
Qwen3-Embedding 4B4B · Q4_K_M3.3 GB61 tok/sEST
Nemotron 3 Nano 4B3.97B · Q4_K_M3.4 GB62 tok/sEST
Ministral 3 3B3.85B · Q5_K_M3.5 GB56 tok/sEST
phi-3-mini-4k 3.8B3.8B · Q4_K_M3.4 GB64 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 R9 290X24 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 R9 290X2 — 4 GB VRAM.

▸ RADEON R9 290X2 SPEC
BRAND
AMD
VRAM
4 GB GDDR5
BANDWIDTH
346 GB/s
FP16 COMPUTE
5.6 TFLOPS
FP32 COMPUTE
5.6 TFLOPS
STREAM PROCESSORS
2,816
TDP
580 W
ARCHITECTURE
GCN 2.0
▸ AI CAPABILITY
97/ 449 models @ Q4

With 4 GB VRAM and 346 GB/s bandwidth, this GPU handles models up to 4.25B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~40 tok/s.

§ 01TOP MODELS FOR RADEON R9 290X2
97 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Ministral 3 3B Reasoning4.25B3.1 GB72—
Qwen 1.5 4B4B2.9 GB7712.6
Qwen3 4B4B2.9 GB7740.7
Qwen3-4B Instruct 25074B2.9 GB7737.2
Qwen3-Embedding 4B4B2.9 GB77—
Nemotron 3 Nano 4B3.97B2.9 GB7732.0
Ministral 3 3B3.85B2.8 GB8021.4
Phi-3.5 Mini 3.8B3.82B2.8 GB8146.6
phi-3-mini-4k 3.8B3.8B2.8 GB8130.5
Phi-4-mini 3.8B3.8B2.8 GB8149.0
Cogito 3B3.61B2.7 GB8522.1
Falcon3-3B3.23B2.5 GB9525.7
granite-4.0-h-micro 3.2B3.2B2.4 GB9618.4
Llama-3.2-3B3.2B2.4 GB9617.9
Falcon-H1 3B3.15B2.4 GB9849.5
Qwen 2.5 3B3.1B2.4 GB9937.2
SmolLM3-3B3.1B2.4 GB9930.5
Ministral 3B3B2.3 GB10329.6
StarCoder2 3B3B2.3 GB1039.5
Granite 4.1 3B3B2.3 GB10316.6