llmrun

NVIDIA · Ada Lovelace · 12 GB VRAM tier

Best AI Models for NVIDIA GeForce RTX 4070 (12 GB)

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

12 GB GDDR6X504 GB/s58 TFLOPS FP16200 WAda Lovelace 2023
Buyabout $599

What LLMs can the RTX 4070 run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 51.5 t/s
51.5 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 101 t/s
101 tok/sQ4_K_M · 3.4 GB
3
Gemma 3 12B ITGoogle
Q4_K_M · 10.9 GB · 40.7 t/s
40.7 tok/sQ4_K_M · 10.9 GB
4
Gemma 2 9B ITGoogle
Q4_K_M · 7.3 GB · 53.7 t/s
53.7 tok/sQ4_K_M · 7.3 GB
5
Ling 3.0 TinyInclusion AI
Q4_K_M · 5.5 GB · 240 t/s
240 tok/sQ4_K_M · 5.5 GB
See all 276 models on the RTX 4070 →Best quant that fits · speeds estimated on this card

About the RTX 4070

The RTX 4070 is a solid mid-range card in the 12 GB class: it runs 7–9B models well. It is more efficient per watt than most 12 GB cards, and pricier per token than most of the cards compared at ≈$575 used. Ada Lovelace generation, released 2023.

Speed in 12 GB#3504 GB/s
Per watt#30.26 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#69.0 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationAda Lovelacereleased 2023

RTX 4070 strengths and weaknesses

Strengths

  • Runs 14B models: Qwen3 14B at 34.5 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
  • Used prices still near MSRP: ≈$575 used vs its $599 MSRP

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 RTX 4070 has no new price, so the New view shows it 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 407012 GB
about $59951.5
8.6
25.8
RTX 507012 GB
about $84068.7
8.2
27.5
RX 7900 XTX24 GB
about $1,13090.6
8.0
25.5
RX 907016 GB
about $80060.4
7.6
27.5
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 507012 GB
about $73668.7
9.3
27.5
RX 907016 GB
about $65060.4
9.3
27.5
RX 7900 XTX24 GB
about $1,00090.6
9.1
25.5
RTX 407012 GB
about $57551.5
9.0
25.8
RTX 4070 SUPER12 GB
about $67551.5
7.6
23.4
RX 9060 XT 16GB16 GB
about $48030.2
6.3
18.9

Cost per token: the RTX 4070 against 8–24 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 GB24 GBthis cardat MSRP
$0.00$0.50$1.00$1.50$0$1,000$2,000$ per 1M output tokens · lower is betterused price →RTX 5070 · 12 GB · ≈$736 · $0.83 per 1M tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,000 · $0.86 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$480 · $1.23 per 1M tokensRTX 5060 · 8 GB · ≈$388 · $0.67 per 1M tokensRTX 5080 · 16 GB · ≈$1,640 · $1.21 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,100 · $0.90 per 1M tokensRTX 4080 SUPER · 16 GB · ≈$1,155 · $1.15 per 1M tokensRTX 4070 Ti SUPER · 16 GB · ≈$900 · $1.00 per 1M tokensRTX 4080 · 16 GB · ≈$1,100 · $1.13 per 1M tokensRX 9070 XT · 16 GB · ≈$790 · $1.04 per 1M tokensRTX 4070 SUPER · 12 GB · ≈$675 · $1.01 per 1M tokensRTX 4070 Ti · 12 GB · ≈$722 · $1.12 per 1M tokensRTX 4060 · 8 GB · ≈$335 · $0.94 per 1M tokensRTX 4060 Ti 8GB · 8 GB · ≈$362 · $1.01 per 1M tokensRTX 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 4070RTX 5060RTX 5080RTX 5070 TiRX 7900 XTXRTX 4080 SUPERRTX 4080
$0.00$0.50$1.00$1.50$0$1k$2k$ per 1M output tokens · lower is betterused price →RTX 5070 · 12 GB · ≈$736 · $0.83 per 1M tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,000 · $0.86 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$480 · $1.23 per 1M tokensRTX 5060 · 8 GB · ≈$388 · $0.67 per 1M tokensRTX 5080 · 16 GB · ≈$1,640 · $1.21 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,100 · $0.90 per 1M tokensRTX 4080 SUPER · 16 GB · ≈$1,155 · $1.15 per 1M tokensRTX 4070 Ti SUPER · 16 GB · ≈$900 · $1.00 per 1M tokensRTX 4080 · 16 GB · ≈$1,100 · $1.13 per 1M tokensRX 9070 XT · 16 GB · ≈$790 · $1.04 per 1M tokensRTX 4070 SUPER · 12 GB · ≈$675 · $1.01 per 1M tokensRTX 4070 Ti · 12 GB · ≈$722 · $1.12 per 1M tokensRTX 4060 · 8 GB · ≈$335 · $0.94 per 1M tokensRTX 4060 Ti 8GB · 8 GB · ≈$362 · $1.01 per 1M tokensRTX 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 4070RTX 5060RTX 5080
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$0.87
Card, used price
$0.71
Power
$0.16
3-year cost of ownership$706
Or rent in the cloud$0.14/hr

Renting costs as much as buying after ≈4,200 hours (≈2.9 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 RTX 4070, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

4 GB more and 17% fasterRX 9070AMD · 16 GB · RDNA 4
▲ 17% speed▲ 4 GBabout $800
12 GB more and 76% fasterRX 7900 XTXAMD · 24 GB · RDNA 3
▲ 76% speed▲ 12 GBabout $1,130
29% slower, same 12 GBRTX 3060 12GBNVIDIA · 12 GB · Ampere
▼ 28% speedsame 12 GBabout $329rent $0.07/hr
Same 12 GB, new with warrantyRTX 5070NVIDIA · 12 GB · Blackwell
▲ 33% speedsame 12 GBabout $840rent $0.14/hr
4 GB more, but slowerRX 9060 XT 16GBAMD · 16 GB · RDNA 4
▼ 41% speed▲ 4 GBabout $590
4 GB more and 14% fasterRX 7800 XTAMD · 16 GB · RDNA 3
▲ 14% speed▲ 4 GBabout $499

RTX 4070 tokens per second by model size

Output speed depends on memory bandwidth (504 GB/s). Grey rows need more than 12 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    51.5tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB tight · Q4_K_M
    34.5tok/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 RTX 4070

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 NVIDIA GeForce RTX 4070 — ranked by Score, recency and Fit
ModelVRAMScoreFit
Qwen2 1.5B Instruct
Q4_K_M·255.9 t/s
1.7 GBS100, Runs great
Vlt5 Base Keywords
BF16·537.0 t/s
1.6 GBS100, Runs great
Devstral Small 2 24B Instruct 2512
IQ2_M·37.1 t/s
11.8 GBC44, Usable
Qwen2.5 0.5B
Q4_K_M·528.4 t/s
0.8 GBS100, Runs great
Yi 9B
Q4_K_M·56.5 t/s
6 GBA82, Runs well
Gemma 2 2B Jpn IT
BF16·57.0 t/s
7.6 GBA82, Runs well
Qwen1.5 1.8B Chat
Q4_K_M·182.0 t/s
4.6 GBS99, Runs great
Llama 3.2 1B
Q4_K_M·399.5 t/s
2.3 GBS100, Runs great
BioMistral 7B
Q4_K_M·68.7 t/s
6.7 GBS85, Runs great
CodeQwen1.5 7B
Q4_K_M·68.5 t/s
5.7 GBS85, Runs great
Qwen2.5 1.5B
Q4_K_M·255.9 t/s
1.7 GBS100, Runs great
Baichuan2 7B Base
Q4_K_M·70.9 t/s
5.0 GBS86, Runs great
Xgen 7B 8k Base
Q4_K_M·70.9 t/s
5.7 GBS86, Runs great
Shieldgemma 2B
Q4_K_M·189.4 t/s
3.6 GBS99, Runs great
SmolLM 360M
Q4_K_M·546.0 t/s
0.6 GBS100, Runs great

RTX 4070 specs, power and PSU

Specs

Architecture
Ada Lovelace
VRAM
12 GB GDDR6X
Memory bandwidth
504.0 GB/s
FP16 (dense)
58.30 TFLOPS
CUDA cores
5,888
Tensor cores
184
Compute capability
8.9
TDP
200 W
Released
2023-04-13
MSRP
$599

RTX 4070 FAQ

Is the RTX 4070 still worth it in 2026?

At ≈$575 used you get 9.0 tok/s per $100 on Qwen3.5 9B, which ranks 6th of the 8 cards compared here; it is also the 3rd fastest out of 9 cards with 12 GB. It runs 7–9B models well; 14B only just fits.

What is the largest model the RTX 4070 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 119 tok/s.

Can the RTX 4070 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 RTX 4070 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 RTX 4070 give?

About 51.5 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 3.0 s on a 4K-token prompt.

What power supply does the RTX 4070 need?

Plan a 400 W PSU (2× the 200 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 62 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an RTX 4070?

Rent from $0.14/hr. Renting costs as much as buying after ≈4,200 hours (≈2.9 years at 4 h a day).

Related

Best coding LLMs12 GB VRAM16 GB VRAMGPUs with 12 GB or moreRTX 3090 TiRTX 4090
Best AI Models for NVIDIA GeForce RTX 4070 (12 GB) | llmrun