llmrun

AMD · RDNA 4 · 12 GB VRAM tier

Best AI Models for AMD Radeon RX 9070 GRE (12 GB)

Runs 7–9B models well; 14B only just fits.

12 GB GDDR6432 GB/s137 TFLOPS FP16220 WRDNA 4 2026
Buyabout $549

12 GB caps what fits locally — a step down from the 16 GB RX 9070/9070 XT. Launched China-only in May 2025; this is the worldwide retail launch.

What LLMs can the RX 9070 GRE run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 40.8 t/s
40.8 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 79.5 t/s
79.5 tok/sQ4_K_M · 3.4 GB
3
Gemma 3 12B ITGoogle
Q4_K_M · 10.9 GB · 32.2 t/s
32.2 tok/sQ4_K_M · 10.9 GB
4
Gemma 2 9B ITGoogle
Q4_K_M · 7.3 GB · 42.5 t/s
42.5 tok/sQ4_K_M · 7.3 GB
5
Ling 3.0 TinyInclusion AI
Q4_K_M · 5.5 GB · 119 t/s
119 tok/sQ4_K_M · 5.5 GB
See all 276 models on the RX 9070 GRE →Best quant that fits · speeds estimated on this card

About the RX 9070 GRE

The RX 9070 GRE is an entry card in the 12 GB class: it runs 7–9B models well. It is power-hungry for its speed, and at its $549 MSRP it cost more per token than most of the cards compared. RDNA 4 generation, released 2026.

Speed in 12 GB#6432 GB/s
Per watt#70.19 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#57.4 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationRDNA 4released 2026

RX 9070 GRE strengths and weaknesses

Strengths

  • Runs 14B models: Qwen3 14B at 27.3 tok/s in 11.9 GB

Weaknesses

  • 27–31B models don't fit: Qwen3.8 27B needs 17.9 GB at Q4 vs 12 GB here
  • Little room for context: Qwen3 14B uses 11.9 of 12 GB, 0.1 GB left
  • ROCm, not CUDA: No FP8 weights; MXFP4 only through llama.cpp, not vLLM

Speed per dollar and per watt

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 9070 GRE has no current price, so it is shown at its MSRP.

CardNew pricetok/stok/s per $100tok/s per 100 W
RTX 50608 GBBest new value
about $46045.8
10.0
31.6
RTX 507012 GB
about $84068.7
8.2
27.5
RTX 5070 Ti16 GB
about $1,17091.6
7.8
30.5
RX 907016 GB
about $80060.4
7.6
27.5
RX 9070 GRE12 GB
about $54940.8
7.4
18.5
RX 9070 XT16 GB
about $81560.4
7.4
19.9
RTX 40608 GB
about $52027.8
5.3
24.2
RX 9060 XT 16GB16 GB
about $59030.2
5.1
18.9
CardUsed pricetok/stok/s per $100tok/s per 100 W
RTX 50608 GBBest used value
about $38845.8
11.8
31.6
RTX 3060 12GB12 GB
about $32936.8
11.2
21.6
RTX 407012 GB
about $57551.5
9.0
25.8
RTX 40608 GB
about $33527.8
8.3
24.2
RTX 4060 Ti 8GB8 GB
about $36229.4
8.1
18.4
RTX 4070 SUPER12 GB
about $67551.5
7.6
23.4
RTX 4070 Ti12 GB
about $72251.5
7.1
18.1
RX 9060 XT 16GB16 GB
about $48030.2
6.3
18.9

Cost per token: the RX 9070 GRE against 8–16 GB cards

On Qwen3.5 9B (9B params, Q4_K_M, 4K context). Speed, cost per token and power per token depend on the model.

8 GB12 GB16 GBthis cardat MSRP
$0.00$0.50$1.00$1.50$2.00$0$1,000$2,000$ per 1M output tokens · lower is betternew price →RX 9060 XT 16GB · 16 GB · ≈$590 · $1.46 per 1M tokensRTX 5060 · 8 GB · ≈$460 · $0.77 per 1M tokensRTX 5080 · 16 GB · ≈$1,690 · $1.25 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,170 · $0.95 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRX 9070 XT · 16 GB · ≈$815 · $1.07 per 1M tokensRX 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRTX 4060 · 8 GB · ≈$520 · $1.36 per 1M tokensRX 9070 GRE · 12 GB · $549 MSRP · $1.08 per 1M tokensRX 9070 GRERTX 5060RTX 5080RTX 5070 TiRTX 5070RX 9070 XTRX 9070
$0.00$0.50$1.00$1.50$2.00$0$1k$2k$ per 1M output tokens · lower is betternew price →RX 9060 XT 16GB · 16 GB · ≈$590 · $1.46 per 1M tokensRTX 5060 · 8 GB · ≈$460 · $0.77 per 1M tokensRTX 5080 · 16 GB · ≈$1,690 · $1.25 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,170 · $0.95 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRX 9070 XT · 16 GB · ≈$815 · $1.07 per 1M tokensRX 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRTX 4060 · 8 GB · ≈$520 · $1.36 per 1M tokensRX 9070 GRE · 12 GB · $549 MSRP · $1.08 per 1M tokensRX 9070 GRERTX 5060RTX 5080
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$1.08
Card, MSRP
$0.85
Power
$0.22
3-year cost of ownership$694
Or rent an RTX 3090 in the cloud$0.17/hr

Renting costs as much as buying after ≈3,200 hours (≈2.2 years at 4 h a day).

Include card price

Each dot is a card; hover or tap it for its numbers, click to open it. Cards without a price for the chosen basis are left out. How we calculate cost per token

Similar GPUs

Cards people weigh against the RX 9070 GRE, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

26% faster, same 12 GBRTX 4070NVIDIA · 12 GB · Ada Lovelace
▲ 26% speedsame 12 GBabout $599rent $0.14/hr
10% slower, same 12 GBRTX 3060 12GBNVIDIA · 12 GB · Ampere
▼ 10% speedsame 12 GBabout $329rent $0.07/hr
26% faster, same 12 GBRTX 4070 SUPERNVIDIA · 12 GB · Ada Lovelace
▲ 26% speedsame 12 GBabout $599rent $0.10/hr
4 GB more, but slowerRX 9060 XT 16GBAMD · 16 GB · RDNA 4
▼ 26% speed▲ 4 GBabout $590
28% slower, 8 GBRTX 4060 Ti 8GBNVIDIA · 8 GB · Ada Lovelace
▼ 28% speed▼ 4 GBabout $399rent $0.26/hr
4 GB more and 8% fasterA770 16GBIntel · 16 GB · Alchemist
▲ 8% speed▲ 4 GBabout $349

RX 9070 GRE tokens per second by model size

Output speed depends on memory bandwidth (432 GB/s). Grey rows need more than 12 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    40.8tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB tight · Q4_K_M
    27.3tok/s
  3. 27–31BQwen3.8 27B · not at 4-bit
    needs 17.9 GB
  4. 35B MoEQwen3.6 35B A3B · not at 4-bit
    needs 22.0 GB
  5. 70BLlama 3.3 70B Instruct · not at 4-bit
    needs 51.3 GB
  6. 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 956 models on the RX 9070 GRE

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 9070 GRE — ranked by Score, recency and Fit
ModelVRAMScoreFit
Qwen3 4B Abliterated
Q4_K_M·97.8 t/s
4.8 GBS91, Runs great
Qwen3 30B A3B Instruct 2507
IQ2_M·70.7 t/s
11.4 GBB62, Decent
Yi 6B
Q4_K_M·63.7 t/s
4.2 GBA84, Runs well
Qwen2.5 14B Instruct
Q4_K_S·28.8 t/s
11.8 GBC40, Usable
Turkish Gemma 9B v0.1
Q4_K_M·40.1 t/s
7.7 GBA75, Runs well
Macbert4csc Base Chinese
BF16·1127.0 t/s
0.2 GBS100, Runs great
Falcon Mamba 7B
Q4_K_M·54.0 t/s
7.3 GBA81, Runs well
Qwen2 1.5B Instruct
Q4_K_M·202.5 t/s
1.7 GBS100, Runs great
Vlt5 Base Keywords
BF16·424.9 t/s
1.6 GBS100, Runs great
Nemotron Mini 4B Instruct
Q4_K_M·87.3 t/s
3.2 GBS89, Runs great
SmolLM 1.7B
Q4_K_M·149.8 t/s
1.7 GBS97, Runs great
Qwen2.5 0.5B
Q4_K_M·418.1 t/s
0.8 GBS100, Runs great
Devstral Small 2 24B Instruct 2512
IQ2_M·29.4 t/s
11.8 GBC42, Usable
Qwen2.5 7B
Q4_K_M·51.9 t/s
5.8 GBA81, Runs well
Mellum 4B Base
Q4_K_M·74.7 t/s
5.7 GBS86, Runs great

RX 9070 GRE specs, power and PSU

Specs

Architecture
RDNA 4
VRAM
12 GB GDDR6
Memory bandwidth
432.0 GB/s
FP16 (dense)
137.20 TFLOPS
Stream processors
3,072
TDP
220 W
Released
2026-06-01
MSRP
$549

RX 9070 GRE FAQ

Is the RX 9070 GRE still worth it in 2026?

At its MSRP of $549 you get 7.4 tok/s per $100 on Qwen3.5 9B, which ranks 5th of the 8 cards compared here; it is also the 6th fastest out of 9 cards with 12 GB. It runs 7–9B models well; 14B only just fits.

What is the largest model the RX 9070 GRE can run?

GPT OSS 20B (20.9B) is the largest at 4-bit or better: at IQ4_XS it uses 11.8 GB of 12 GB, and runs at about 72.6 tok/s.

Can the RX 9070 GRE run Qwen3.5 9B?

Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of 12 GB, and it is the best model that runs well on this card.

Can the RX 9070 GRE run 27–31B models?

No. Qwen3.8 27B needs 17.9 GB at Q4, more than the 12 GB on this card.

How many tokens per second does the RX 9070 GRE give?

About 40.8 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 2.6 s on a 4K-token prompt.

What power supply does the RX 9070 GRE need?

Plan a 450 W PSU (2× the 220 W TDP).

Is 12 GB enough for coding and agents?

Yes for coding: Llama 3.1 8B Instruct, the only coding-scored model that fits, runs at about 49 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an RX 9070 GRE?

An RTX 3090 (24 GB) rents from $0.17/hr. Renting costs as much as buying after ≈3,200 hours (≈2.2 years at 4 h a day).

Related

Best coding LLMs12 GB VRAM16 GB VRAMGPUs with 12 GB or moreRTX 3090 TiRTX 4090