Best AI Models for AMD Radeon RX 5700 XT (8 GB)

Runs 7–9B models well; 14B doesn't fit in 8 GB.

8 GB GDDR6448 GB/s20 TFLOPS FP16225 WRDNA 2019

8 GB at 448 GB/s runs ~7-8B Q4 models at a usable speed. AMD's ROCm and Ollama don't support RX 5000 cards; run llama.cpp or LM Studio with the Vulkan backend.

What LLMs can the RX 5700 XT run?

#	Model	Speed	Quant · VRAM
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 42.3 t/s	42.3 tok/s	Q4_K_M · 6.6 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 82.5 t/s	82.5 tok/s	Q4_K_M · 3.4 GB
3	Gemma 2 9B ITGoogle Q4_K_M · 7.3 GB · 44.1 t/s	44.1 tok/s	Q4_K_M · 7.3 GB
4	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 94.6 t/s	94.6 tok/s	Q4_K_M · 5.0 GB
5	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 121 t/s	121 tok/s	Q4_K_M · 5.5 GB

#	Model	Speed	Reasoning Score	Quant · VRAM
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 42.3 t/s	42.3 tok/s		Q4_K_M · 6.6 GB
2	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 94.6 t/s	94.6 tok/s		Q4_K_M · 5.0 GB
3	Qwen2.5 7B InstructAlibaba Q4_K_M · 5.8 GB · 53.9 t/s	53.9 tok/s		Q4_K_M · 5.8 GB
4	Gemma 2 9B ITGoogle Q4_K_M · 7.3 GB · 44.1 t/s	44.1 tok/s	overall	Q4_K_M · 7.3 GB
5	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 121 t/s	121 tok/s		Q4_K_M · 5.5 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 82.5 t/s	82.5 tok/s	Q4_K_M · 3.4 GB
2	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 42.3 t/s	42.3 tok/s	Q4_K_M · 6.6 GB
3	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 50.5 t/s	50.5 tok/s	Q4_K_M · 5.6 GB
4	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 121 t/s	121 tok/s	Q4_K_M · 5.5 GB
5	Spark X2.6 4BXHToken Q4_K_M · 4.3 GB · 90.8 t/s	90.8 tok/s	Q4_K_M · 4.3 GB

#	Model	Speed ↓	Quant · VRAM
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 82.5 t/s	82.5 tok/s	Q4_K_M · 3.4 GB
2	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 121 t/s	121 tok/s	Q4_K_M · 5.5 GB
3	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 50.5 t/s	50.5 tok/s	Q4_K_M · 5.6 GB
4	MiMo V2.6 Distill Qwen 9BXiaomi Q4_K_M · 6.5 GB · 43.3 t/s	43.3 tok/s	Q4_K_M · 6.5 GB
5	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 42.3 t/s	42.3 tok/s	Q4_K_M · 6.6 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 42.3 t/s	42.3 tok/s	Q4_K_M · 6.6 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 82.5 t/s	82.5 tok/s	Q4_K_M · 3.4 GB
3	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 94.6 t/s	94.6 tok/s	Q4_K_M · 5.0 GB
4	MiMo V2.6 Distill Qwen 9BXiaomi Q4_K_M · 6.5 GB · 43.3 t/s	43.3 tok/s	Q4_K_M · 6.5 GB
5	MiniCPM V 4.7 1BOpenBMB Q4_K_M · 1.6 GB · 210 t/s	210 tok/s	Q4_K_M · 1.6 GB

See all 218 models on the RX 5700 XT →Best quant that fits · speeds estimated on this card

About the RX 5700 XT

The RX 5700 XT is an entry card in the 8 GB class: it runs 7–9B models well. It is power-hungry for its speed, and the best speed per dollar of the cards compared at ≈$169 used. RDNA generation, released 2019.

Speed in 8 GB#11448 GB/s

Per watt#140.19 tok/s per W on Qwen3.5 9B (9B params)

Per dollar#125.0 tok/s per $100 on Qwen3.5 9B (9B params)

GenerationRDNAreleased 2019

RX 5700 XT strengths and weaknesses

Strengths

Runs 7–9B models: Qwen3.5 9B at 42.3 tok/s in 6.6 GB

Weaknesses

14B models don't fit: Qwen3 14B needs 11.9 GB at Q4 vs 8 GB here
Little room for context: Qwen3.5 9B uses 6.6 of 8 GB, 1.4 GB left
Slow prompt reading: A 4K-token prompt on Qwen3.5 9B waits 18.5 s for the first token
ROCm, not CUDA: No FP8 weights; MXFP4 only through llama.cpp, not vLLM

Speed per dollar and per watt

NewUsed

On Qwen3.5 9B (9B params) at Q4_K_M. New prices are compared with new, used with used. The last two columns are the speed you get per $100 of price and per 100 W of board power; higher is better. The RX 5700 XT has no new price, so the New view shows it at its MSRP.

Card	New price	tok/s	tok/s per $100	tok/s per 100 W
RX 5700 XT8 GB	about $399	42.3	10.6	18.8
RTX 50608 GBBest new value	about $460	45.8	10.0	31.6
RTX 507012 GB	about $840	68.7	8.2	27.5
RTX 40608 GB	about $520	27.8	5.3	24.2

Card	Used price	tok/s	tok/s per $100	tok/s per 100 W
RX 5700 XT8 GB	about $169	42.3	25.0	18.8
RTX 3060 12GB12 GB	about $329	36.8	11.2	21.6
RTX 50608 GBBest used value	about $430	45.8	10.7	31.6
RTX 507012 GB	about $736	68.7	9.3	27.5
RTX 407012 GB	about $575	51.5	9.0	25.8
RTX 40608 GB	about $335	27.8	8.3	24.2
RTX 4070 SUPER12 GB	about $675	51.5	7.6	23.4
RTX 4070 Ti12 GB	about $722	51.5	7.1	18.1

Similar GPUs

Cards people weigh against the RX 5700 XT, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

8% faster, same 8 GBRTX 5060NVIDIA · 8 GB · Blackwell

▲ 8% speedsame 8 GB$460

4 GB more and 62% fasterRTX 5070NVIDIA · 12 GB · Blackwell

▲ 62% speed▲ 4 GB$840rent $0.15/hr

Same 8 GB, new with warrantyRTX 4060NVIDIA · 8 GB · Ada Lovelace

▼ 34% speedsame 8 GB$520rent $0.08/hr

4 GB more, but slowerRTX 3060 12GBNVIDIA · 12 GB · Ampere

▼ 13% speed▲ 4 GB$329rent $0.06/hr

About as fast, same 8 GBA580Intel · 8 GB · Alchemist

▼ 5% speedsame 8 GB$179

8 GB more and 73% fasterRTX 4080NVIDIA · 16 GB · Ada Lovelace

▲ 73% speed▲ 8 GB$1,199rent $0.24/hr

RX 5700 XT tokens per second by model size

Output speed depends on memory bandwidth (448 GB/s). Grey rows need more than 8 GB.

9BQwen3.5 9B · Q4_K_M · 6.6 GB
42.3tok/s
14BQwen3 14B · not at 4-bit
needs 11.9 GB
27–31BQwen3.8 27B · not at 4-bit
needs 17.9 GB
35B MoEQwen3.6 35B A3B · not at 4-bit
needs 22.0 GB
70BLlama 3.3 70B Instruct · not at 4-bit
needs 51.3 GB
120B MoEGPT OSS 120B · not at 4-bit
needs 70.8 GB

Estimated speeds: Q4_K_M, 4K context, one user. How we estimate

All 842 models on the RX 5700 XT

Featured only

Best quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with AMD Radeon RX 5700 XT — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Param 1 2.9B Instruct2.9BChat BF16·42.7 t/s	BF16	7.1 GB89%	42.7 t/s	8K	B64, Decent
Qwen3 Code Reasoning 4B4BChatCodeReasoning Q4_K_M·93.0 t/s	Q4_K_M	4.2 GB53%	93.0 t/s	262K	S90, Runs great
Mamba 130M HF129MChat Q4_K_M·2986.7 t/s	Q4_K_M	0.9 GB11%	2986.7 t/s	—	S100, Runs great
Tiny Random PhiForCausalLM80074Chat Q4_K_M·896.0 t/s	Q4_K_M	0.3 GB4%	896.0 t/s	1K	S100, Runs great
Yi 1.5 6B Chat6.1BChat Q4_K_M·66.0 t/s	Q4_K_M	4.2 GB53%	66.0 t/s	4K	S85, Runs great
Dobby Mini Unhinged Llama 3.1 8B8.0BChatRoleplay Q4_K_M·49.9 t/s	Q4_K_M	7.3 GB91%	49.9 t/s	131K	B63, Decent
Bloomz 560M559MChat Q4_K_M·726.5 t/s	Q4_K_M	0.4 GB5%	726.5 t/s	2K	S100, Runs great
Bloom 560M559MChat Q4_K_M·726.5 t/s	Q4_K_M	1.6 GB20%	726.5 t/s	—	S100, Runs great
Qwen3 8B8BChat Q4_K_M·50.9 t/s	Q4_K_M	7.8 GB98%	50.9 t/s	—	C44, Usable
Goedel Prover v2 8B8.2BChat Q4_K_M·48.7 t/s	Q4_K_M	7.6 GB95%	48.7 t/s	41K	B56, Decent
Pythia 70M Deduped96MChat Q4_K_M·4480.0 t/s	Q4_K_M	0.1 GB1%	4480.0 t/s	2K	S100, Runs great
Idefics3 8B Llama38.5BVision Q4_K_M·47.6 t/s	Q4_K_M	7.5 GB94%	47.6 t/s	131K	B58, Decent
Japanese GPT Neox Small204MChat Q4_K_M·2067.7 t/s	Q4_K_M	0.1 GB2%	2067.7 t/s	2K	S100, Runs great
GPT Neo 125M150MChat BF16·814.5 t/s	BF16	0.3 GB4%	814.5 t/s	2K	S100, Runs great
Gemma 3n E2B IT5.4BVision Q4_K_M·74.9 t/s	Q4_K_M	5.9 GB74%	74.9 t/s	—	S86, Runs great

RX 5700 XT specs, power and PSU

Specs

Architecture: RDNA
VRAM: 8 GB GDDR6
Memory bandwidth: 448.0 GB/s
FP16 (dense): 19.51 TFLOPS
Stream processors: 2,560
TDP: 225 W
Released: 2019-07-07
MSRP: $399

RX 5700 XT FAQ

Is the RX 5700 XT still worth it in 2026?

At ≈$169 used you get 25.0 tok/s per $100 on Qwen3.5 9B, which ranks 1st of the 8 cards compared here; it is also the 11th fastest out of 20 cards with 8 GB. It runs 7–9B models well; 14B doesn't fit in 8 GB.

What is the largest model the RX 5700 XT can run?

Qwen3.5 9B (9.7B) is the largest at 4-bit or better: at Q4_K_M it uses 6.6 GB of 8 GB, and runs at about 42.3 tok/s.

Can the RX 5700 XT run 14B models?

No. Qwen3 14B needs 11.9 GB at Q4, more than the 8 GB on this card.

How many tokens per second does the RX 5700 XT give?

About 42.3 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 18.4 s on a 4K-token prompt.

What power supply does the RX 5700 XT need?

Plan a 450 W PSU (2× the 225 W TDP).

Is renting cheaper than buying an RX 5700 XT?

An RTX 3090 (24 GB) rents from $0.18/hr. Renting costs as much as buying after ≈900 hours (≈0.6 years at 4 h a day).

Best coding LLMs 8 GB VRAM 12 GB VRAM GPUs with 8 GB or more RTX 5080 RTX 3080 Ti