llmrun

NVIDIA · Pascal

Best AI Models for NVIDIA GeForce GTX 1080 Ti (11 GB)

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

11 GB GDDR5X484 GB/s11 TFLOPS FP16250 WPascal 2017
Buyabout $699

What LLMs can the GTX 1080 Ti run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 49.5 t/s
49.5 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 96.6 t/s
96.6 tok/sQ4_K_M · 3.4 GB
3
Gemma 2 9B ITGoogle
Q4_K_M · 7.3 GB · 51.6 t/s
51.6 tok/sQ4_K_M · 7.3 GB
4
Gemma 3 4B ITGoogle
Q4_K_M · 5.0 GB · 111 t/s
111 tok/sQ4_K_M · 5.0 GB
5
Ling 3.0 TinyInclusion AI
Q4_K_M · 5.5 GB · 233 t/s
233 tok/sQ4_K_M · 5.5 GB
See all 265 models on the GTX 1080 Ti →Best quant that fits · speeds estimated on this card

About the GTX 1080 Ti

The GTX 1080 Ti is an entry card in the 10–11 GB class: it runs 7–9B models well. It is power-hungry for its speed, and at its $699 MSRP it cost more per token than most of the cards compared. Pascal generation, released 2017.

Speed in 10–11 GB#3484 GB/s
Per watt#40.20 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#57.1 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationPascalreleased 2017

GTX 1080 Ti strengths and weaknesses

Strengths

  • Runs 14B models: Qwen3 14B at 36.7 tok/s in 10.9 GB

Weaknesses

  • 27–31B models don't fit: Qwen3.8 27B needs 17.9 GB at Q4 vs 11 GB here
  • Little room for context: Qwen3 14B uses 10.9 of 11 GB, 0.1 GB left
  • Slow prompt reading: A 4K-token prompt on Qwen3.5 9B waits 15.5 s for the first token
  • No FP8 or NVFP4: Pascal predates both; use GGUF, AWQ or GPTQ builds

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 GTX 1080 Ti 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
RX 7900 XTX24 GB
about $1,13090.6
8.0
25.5
RX 907016 GB
about $80060.4
7.6
27.5
GTX 1080 Ti11 GB
about $69949.5
7.1
19.8
RTX 40608 GB
about $52027.8
5.3
24.2
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
RTX 4070 Ti12 GB
about $72251.5
7.1
18.1

Cost per token: the GTX 1080 Ti 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 GB11 GB16 GB24 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 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,130 · $0.95 per 1M tokensRTX 5060 · 8 GB · ≈$460 · $0.77 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRTX 4060 · 8 GB · ≈$520 · $1.36 per 1M tokensGTX 1080 Ti · 11 GB · $699 MSRP · $1.11 per 1M tokensGTX 1080 TiRTX 5060RX 7900 XTXRTX 5070RTX 4060
$0.00$0.50$1.00$1.50$2.00$0$1k$2k$ per 1M output tokens · lower is betternew price →RX 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,130 · $0.95 per 1M tokensRTX 5060 · 8 GB · ≈$460 · $0.77 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRTX 4060 · 8 GB · ≈$520 · $1.36 per 1M tokensGTX 1080 Ti · 11 GB · $699 MSRP · $1.11 per 1M tokensGTX 1080 TiRTX 5060RX 7900 XTX
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$1.11
Card, MSRP
$0.90
Power
$0.21
3-year cost of ownership$863
Or rent an RTX 3090 in the cloud$0.19/hr

Renting costs as much as buying after ≈3,700 hours (≈2.5 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 GTX 1080 Ti, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

5 GB more and 22% fasterRX 9070AMD · 16 GB · RDNA 4
▲ 22% speed▲ 5 GBabout $800
13 GB more and 83% fasterRX 7900 XTXAMD · 24 GB · RDNA 3
▲ 83% speed▲ 13 GBabout $1,130
1 GB more, but slowerRTX 3060 12GBNVIDIA · 12 GB · Ampere
▼ 26% speed▲ 1 GBabout $329rent $0.06/hr
5 GB more and 39% fasterRTX 4070 Ti SUPERNVIDIA · 16 GB · Ada Lovelace
▲ 39% speed▲ 5 GBabout $799rent $0.15/hr
Best value near the GTX 1080 TiRTX 5070NVIDIA · 12 GB · Blackwell
▲ 39% speed▲ 1 GBabout $840rent $0.15/hr
5 GB more and 48% fasterRTX 4080NVIDIA · 16 GB · Ada Lovelace
▲ 48% speed▲ 5 GBabout $1,199rent $0.24/hr

GTX 1080 Ti tokens per second by model size

Output speed depends on memory bandwidth (484 GB/s). Grey rows need more than 11 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    49.5tok/s
  2. 14BQwen3 14B · IQ4_XS · 10.9 GB tight · IQ4_XS
    36.7tok/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 936 models on the GTX 1080 Ti

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 GTX 1080 Ti — ranked by Score, recency and Fit
ModelVRAMScoreFit
Nemotron Terminal 14B
Q3_K_M·40.2 t/s
10.2 GBB56, Decent
Llama 7B
Q4_K_M·70.8 t/s
4.5 GBS86, Runs great
NVIDIA Nemotron Nano 9B v2
Q4_K_M·52.1 t/s
6.0 GBA81, Runs well
Nemotron Orchestrator 8B
Q4_K_M·57.0 t/s
7.6 GBA82, Runs well
Olmo 3 1025 7B
Q4_K_M·54.8 t/s
8.4 GBA82, Runs well
Llama 3 Korean Bllossom 8B
Q4_K_M·58.4 t/s
6.2 GBA83, Runs well
Amber
Q4_K_M·58.1 t/s
5.4 GBA83, Runs well
Falcon H1 0.5B Base
Q4_K_M·456.3 t/s
1.2 GBS100, Runs great
Qwen2.5 Omni 7B
Q4_K_M·44.5 t/s
9.9 GBB63, Decent
Vicuna 13B V1.3
Q4_K_M·36.7 t/s
8.6 GBA73, Runs well
GPT J 6B
Q4_K_M·79.5 t/s
4.0 GBS88, Runs great
SOLAR 10.7B Instruct v1.0
Q4_K_M·44.1 t/s
7.5 GBA77, Runs well
Granite 3.0 1B A400m Instruct
Q4_K_M·393.1 t/s
1.3 GBS100, Runs great
Starcoder2 3B
Q4_K_M·144.4 t/s
2.6 GBS96, Runs great
DeepSeek R1 Distill Llama 8B
Q4_K_M·58.4 t/s
7.3 GBA83, Runs well

GTX 1080 Ti specs, power and PSU

Specs

Architecture
Pascal
VRAM
11 GB GDDR5X
Memory bandwidth
484.4 GB/s
FP16 (dense)
11.34 TFLOPS
CUDA cores
3,584
Tensor cores
0
Compute capability
6.1
TDP
250 W
Released
2017-03-10
MSRP
$699

GTX 1080 Ti FAQ

Is the GTX 1080 Ti still worth it in 2026?

At its MSRP of $699 you get 7.1 tok/s per $100 on Qwen3.5 9B, which ranks 5th of the 6 cards compared here; it is also the 3rd fastest out of 4 cards with 10–11 GB. It runs 7–9B models well; 14B only just fits.

What is the largest model the GTX 1080 Ti can run?

Phi 4 (14.7B) is the largest at 4-bit or better: at Q4_0 it uses 11.0 GB of 11 GB, and runs at about 39.1 tok/s.

Can the GTX 1080 Ti run Qwen3.5 9B?

Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of 11 GB, and it is the best model that runs well on this card.

Can the GTX 1080 Ti run 27–31B models?

No. Qwen3.8 27B needs 17.9 GB at Q4, more than the 11 GB on this card.

How many tokens per second does the GTX 1080 Ti give?

About 49.5 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 15.5 s on a 4K-token prompt.

What power supply does the GTX 1080 Ti need?

Plan a 500 W PSU (2× the 250 W TDP).

Is 11 GB enough for coding and agents?

Yes for coding: Llama 3.1 8B Instruct, the only coding-scored model that fits, runs at about 59 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying a GTX 1080 Ti?

An RTX 3090 (24 GB) rents from $0.19/hr. Renting costs as much as buying after ≈3,700 hours (≈2.5 years at 4 h a day).

Related

Best coding LLMs8 GB VRAM12 GB VRAMGPUs with 11 GB or moreRTX 3080 TiRTX 5070 Ti