llmrun

NVIDIA · Ada Lovelace · 16 GB VRAM tier

Best AI Models for NVIDIA GeForce RTX 4080 (16 GB)

Runs 14B models well; 27–31B only just fits.

16 GB GDDR6X717 GB/s98 TFLOPS FP16320 WAda Lovelace 2022
Buyabout $1,199

What LLMs can the RTX 4080 run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 73.3 t/s
73.3 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 143 t/s
143 tok/sQ4_K_M · 3.4 GB
3
GPT OSS 20BOpenAI
Q4_K_M · 13.1 GB · 145 t/s
145 tok/sQ4_K_M · 13.1 GB
4
Phi 4Microsoft
Q4_K_M · 12.5 GB · 48.9 t/s
48.9 tok/sQ4_K_M · 12.5 GB
5
Gemma 4 12B ITGoogle
Q4_K_M · 9.0 GB · 56.6 t/s
56.6 tok/sQ4_K_M · 9.0 GB
See all 307 models on the RTX 4080 →Best quant that fits · speeds estimated on this card

About the RTX 4080

The RTX 4080 is a solid mid-range card in the 16 GB class: it runs 14B models well. It is more efficient per watt than most 16 GB cards, and pricier per token than most of the cards compared at ≈$1,100 used. Ada Lovelace generation, released 2022.

Speed in 16 GB#4717 GB/s
Per watt#70.23 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#66.7 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationAda Lovelacereleased 2022

RTX 4080 strengths and weaknesses

Strengths

  • Runs 35B MoE models: Qwen3.6 35B A3B at 151 tok/s in 15.7 GB
  • Reads long prompts fast: A 4K-token prompt on Qwen3.5 9B starts answering in 1.8 s

Weaknesses

  • 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 16 GB here
  • Used prices still near MSRP: ≈$1,100 used vs its $1,199 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 4080 has no new price, so the New view shows it at its MSRP.

CardNew pricetok/stok/s per $100tok/s per 100 W
RTX 507012 GB
about $84068.7
8.2
27.5
RX 7900 XTX24 GB
about $1,13090.6
8.0
25.5
RTX 5070 Ti16 GBBest new value
about $1,17091.6
7.8
30.5
RX 907016 GB
about $80060.4
7.6
27.5
RX 9070 XT16 GB
about $81560.4
7.4
19.9
RTX 408016 GB
about $1,19973.3
6.1
22.9
RTX 508016 GB
about $1,69098.1
5.8
27.3
RX 9060 XT 16GB16 GB
about $59030.2
5.1
18.9
CardUsed pricetok/stok/s per $100tok/s per 100 W
RTX 3060 12GB12 GBBest used value
about $32936.8
11.2
21.6
RX 907016 GB
about $65060.4
9.3
27.5
RX 7900 XTX24 GB
about $1,00090.6
9.1
25.5
RTX 5070 Ti16 GB
about $1,10091.6
8.3
30.5
RTX 4070 Ti SUPER16 GB
about $90068.7
7.6
24.1
RTX 408016 GB
about $1,10073.3
6.7
22.9
RTX 4080 SUPER16 GB
about $1,15575.2
6.5
23.5
RX 9060 XT 16GB16 GB
about $48030.2
6.3
18.9

Cost per token: the RTX 4080 against 12–32 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.

12 GB16 GB24 GB32 GBthis cardat MSRP
$0.00$0.50$1.00$1.50$2.00$2.50$3.00$0$2,000$4,000$6,000$8,000$ per 1M output tokens · lower is betterused price →RTX 5070 Ti · 16 GB · ≈$1,100 · $0.90 per 1M tokensRTX 4070 Ti SUPER · 16 GB · ≈$900 · $1.00 per 1M tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,000 · $0.86 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$480 · $1.23 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRTX 4090 · 24 GB · ≈$3,000 · $2.03 per 1M tokensRTX 5090 · 32 GB · ≈$6,400 · $2.35 per 1M tokensRTX 5080 · 16 GB · ≈$1,640 · $1.21 per 1M tokensRTX 3090 · 24 GB · ≈$1,500 · $1.15 per 1M tokensRTX 4080 SUPER · 16 GB · ≈$1,155 · $1.15 per 1M tokensRTX 5070 · 12 GB · ≈$736 · $0.83 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 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 3090 Ti · 24 GB · ≈$1,600 · $1.17 per 1M tokensRTX 4080 · 16 GB · ≈$1,100 · $1.13 per 1M tokensRTX 4080RTX 5070 TiRTX 5090RTX 4090RTX 3090 TiRX 7900 XTX
$0.00$0.50$1.00$1.50$2.00$2.50$3.00$0$2k$4k$6k$8k$ per 1M output tokens · lower is betterused price →RTX 5070 Ti · 16 GB · ≈$1,100 · $0.90 per 1M tokensRTX 4070 Ti SUPER · 16 GB · ≈$900 · $1.00 per 1M tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,000 · $0.86 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$480 · $1.23 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRTX 4090 · 24 GB · ≈$3,000 · $2.03 per 1M tokensRTX 5090 · 32 GB · ≈$6,400 · $2.35 per 1M tokensRTX 5080 · 16 GB · ≈$1,640 · $1.21 per 1M tokensRTX 3090 · 24 GB · ≈$1,500 · $1.15 per 1M tokensRTX 4080 SUPER · 16 GB · ≈$1,155 · $1.15 per 1M tokensRTX 5070 · 12 GB · ≈$736 · $0.83 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 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 3090 Ti · 24 GB · ≈$1,600 · $1.17 per 1M tokensRTX 4080 · 16 GB · ≈$1,100 · $1.13 per 1M tokensRTX 4080RTX 5070 TiRTX 5090
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$1.13
Card, used price
$0.95
Power
$0.18
3-year cost of ownership$1,310
Or rent in the cloud$0.24/hr

Renting costs as much as buying after ≈4,500 hours (≈3.1 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 4080, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

25% faster, same 16 GBRTX 5070 TiNVIDIA · 16 GB · Blackwell
▲ 25% speedsame 16 GBabout $1,170rent $0.16/hr
6% slower, same 16 GBRTX 4070 Ti SUPERNVIDIA · 16 GB · Ada Lovelace
▼ 6% speedsame 16 GBabout $799rent $0.15/hr
8 GB more and 24% fasterRX 7900 XTXAMD · 24 GB · RDNA 3
▲ 24% speed▲ 8 GBabout $1,130
59% slower, same 16 GBRX 9060 XT 16GBAMD · 16 GB · RDNA 4
▼ 59% speedsame 16 GBabout $590
Same 16 GB, new with warrantyRX 9070AMD · 16 GB · RDNA 4
▼ 18% speedsame 16 GBabout $800
About as fast, same 16 GBRTX 4080 SUPERNVIDIA · 16 GB · Ada Lovelace
same speedsame 16 GBabout $999rent $0.23/hr

RTX 4080 tokens per second by model size

Output speed depends on memory bandwidth (717 GB/s). Grey rows need more than 16 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    73.3tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB
    49.0tok/s
  3. 27–31BQwen3.8 27B · Q3_K_L · 15.4 GB tight · Q3_K_L
    31.1tok/s
  4. 35B MoEQwen3.6 35B A3B · IQ3_S · 15.7 GB tight · IQ3_S
    151tok/s
  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 1057 models on the RTX 4080

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 4080 — ranked by Score, recency and Fit
ModelVRAMScoreFit
SmolLM2 1.7B
Q4_K_M·269.3 t/s
2.9 GBS100, Runs great
Qwen2.5 Math 1.5B
Q4_K_M·364.0 t/s
1.3 GBS100, Runs great
Codegemma 7B IT
Q4_K_M·82.8 t/s
8.2 GBS88, Runs great
MagicPrompt Stable Diffusion
Q4_K_M·5176.9 t/s
0.1 GBS100, Runs great
Yi 1.5 34B Chat
IQ3_S·30.2 t/s
15.9 GBC39, Usable
Deepseek Llm 7B Base
Q4_K_M·84.6 t/s
6.5 GBS88, Runs great
Gemma 2B IT
Q4_K_M·282.4 t/s
3.5 GBS100, Runs great
Hermes 4.3 36B
IQ2_M·36.5 t/s
14.7 GBB56, Decent
Qwen 14B Chat
Q4_K_M·49.8 t/s
10.6 GBA79, Runs well
Tiny Random LlamaForCausalLM
Q4_K_M·46592.0 t/s
0.2 GBS100, Runs great
Hermes 2 Pro Llama 3 8B
Q4_K_M·86.4 t/s
6.2 GBS88, Runs great
Deepseek Coder 1.3B Instruct
Q4_K_M·308.6 t/s
4.3 GBS100, Runs great
Internlm Chat 20B
Q3_K_M·37.1 t/s
15.1 GBB53, Decent
Qwen3.6 35B A3B Claude 4.7 Opus Reasoning Distilled
Q2_K·150.7 t/s
15.7 GBC44, Usable
Qwen2.5 Coder 3B
Q4_K_M·208.9 t/s
2.8 GBS100, Runs great

RTX 4080 specs, power and PSU

Specs

Architecture
Ada Lovelace
VRAM
16 GB GDDR6X
Memory bandwidth
716.8 GB/s
FP16 (dense)
97.50 TFLOPS
CUDA cores
9,728
Tensor cores
304
Compute capability
8.9
TDP
320 W
Released
2022-11-16
MSRP
$1,199

RTX 4080 FAQ

Is the RTX 4080 still worth it in 2026?

At ≈$1,100 used you get 6.7 tok/s per $100 on Qwen3.5 9B, which ranks 6th of the 8 cards compared here; it is also the 4th fastest out of 17 cards with 16 GB. It runs 14B models well; 27–31B only just fits.

What is the largest model the RTX 4080 can run?

Gemma 4 26B A4B IT (25.8B) is the largest at 4-bit or better: at Q4_K_S it uses 15.5 GB of 16 GB, and runs at about 139 tok/s. At 3-bit, a bigger model fits too: Gemma 2 27B IT (27.2B) at IQ3_M, 15.0 GB.

Can the RTX 4080 run Qwen3.5 9B?

Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of 16 GB, and it is the best model that runs well on this card.

Can the RTX 4080 run 70B models?

No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 16 GB on this card.

How many tokens per second does the RTX 4080 give?

About 73.3 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 1.8 s on a 4K-token prompt.

What power supply does the RTX 4080 need?

Plan a 650 W PSU (2× the 320 W TDP).

Is 16 GB enough for coding and agents?

Yes for coding: GPT OSS 20B, the best of the 2 coding models that fit, runs at about 145 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an RTX 4080?

Rent from $0.24/hr. Renting costs as much as buying after ≈4,500 hours (≈3.1 years at 4 h a day).

Related

Best coding LLMs16 GB VRAM24 GB VRAMGPUs with 16 GB or moreRTX 5090RTX 3090 Ti