llmrun

NVIDIA · Ampere

Best AI Models for NVIDIA GeForce RTX 3080 (10 GB)

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

10 GB GDDR6X760 GB/s60 TFLOPS FP16320 WAmpere 2020
Buyabout $699

What LLMs can the RTX 3080 run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 77.7 t/s
77.7 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 152 t/s
152 tok/sQ4_K_M · 3.4 GB
3
Gemma 2 9B ITGoogle
Q4_K_M · 7.3 GB · 81.0 t/s
81.0 tok/sQ4_K_M · 7.3 GB
4
Gemma 3 4B ITGoogle
Q4_K_M · 5.0 GB · 174 t/s
174 tok/sQ4_K_M · 5.0 GB
5
MiMo V2.6 Distill Qwen 9BXiaomi
Q4_K_M · 6.5 GB · 79.6 t/s
79.6 tok/sQ4_K_M · 6.5 GB
See all 244 models on the RTX 3080 →Best quant that fits · speeds estimated on this card

About the RTX 3080

The RTX 3080 is the quickest card in the 10–11 GB class: it runs 7–9B models well. It is among the most efficient per watt in its tier, and at its $699 MSRP it gave the best speed per dollar of the cards compared. Ampere generation, released 2020.

Speed in 10–11 GB#1760 GB/s
Per watt#20.24 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#111.1 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationAmperereleased 2020

RTX 3080 strengths and weaknesses

Strengths

  • Runs 14B models: Qwen3 14B at 67.9 tok/s in 9.6 GB

Weaknesses

  • 27–31B models don't fit: Qwen3.8 27B needs 17.9 GB at Q4 vs 10 GB here
  • Little room for context: Qwen3 14B uses 9.6 of 10 GB, 0.4 GB left
  • No FP8 or NVFP4: Ampere 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 RTX 3080 has no current price, so it is shown at its MSRP.

CardNew pricetok/stok/s per $100tok/s per 100 W
RTX 308010 GB
about $69977.7
11.1
24.3
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
RTX 40608 GB
about $52027.8
5.3
24.2
RTX 509032 GB
about $5,376183
3.4
31.8
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 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
RTX 509032 GB
about $6,400183
2.9
31.8

Cost per token: the RTX 3080 against 8–32 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.

8 GB10 GB24 GB32 GBthis cardat MSRP
$0.00$0.50$1.00$1.50$2.00$2.50$0$2,000$4,000$6,000$ per 1M output tokens · lower is betternew price →RTX 5090 · 32 GB · ≈$5,376 · $1.99 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 tokensRTX 3080 · 10 GB · $699 MSRP · $0.74 per 1M tokensRTX 3080RTX 5060RTX 5090RX 7900 XTXRTX 5070RTX 4060
$0.00$0.50$1.00$1.50$2.00$2.50$0$2k$4k$6k$ per 1M output tokens · lower is betternew price →RTX 5090 · 32 GB · ≈$5,376 · $1.99 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 tokensRTX 3080 · 10 GB · $699 MSRP · $0.74 per 1M tokensRTX 3080RTX 5090RX 7900 XTX
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$0.74
Card, MSRP
$0.57
Power
$0.17
3-year cost of ownership$909
Or rent in the cloud$0.14/hr

Renting costs as much as buying after ≈4,900 hours (≈3.3 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 3080, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

14 GB more and 17% fasterRX 7900 XTXAMD · 24 GB · RDNA 3
▲ 17% speed▲ 14 GBabout $1,130
22 GB more and 2.4× the speedRTX 5090NVIDIA · 32 GB · Blackwell
▲ 136% speed▲ 22 GBabout $5,376rent $0.50/hr
2 GB more, but slowerRTX 3060 12GBNVIDIA · 12 GB · Ampere
▼ 53% speed▲ 2 GBabout $329rent $0.06/hr
10 GB more for bigger modelsRX 7900 XTAMD · 20 GB · RDNA 3
same speed▲ 10 GBabout $899
6 GB more for bigger modelsRTX 4080 SUPERNVIDIA · 16 GB · Ada Lovelace
▼ 3% speed▲ 6 GBabout $999rent $0.26/hr
6 GB more for bigger modelsRTX 4080NVIDIA · 16 GB · Ada Lovelace
▼ 6% speed▲ 6 GBabout $1,199rent $0.27/hr

RTX 3080 tokens per second by model size

Output speed depends on memory bandwidth (760 GB/s). Grey rows need more than 10 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    77.7tok/s
  2. 14BQwen3 14B · IQ3_M · 9.6 GB tight · IQ3_M
    67.9tok/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 903 models on the RTX 3080

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 3080 — ranked by Score, recency and Fit
ModelVRAMScoreFit
MiniCPM V 4.7 1B
Q4_K_M·386.1 t/s
1.6 GBS100, Runs great
LFM3 VL 2B
Q4_K_M·318.8 t/s
2.3 GBS100, Runs great
Clef Flash
Q4_K_M·79.6 t/s
6.5 GBS88, Runs great
D1 3B
Q4_K_M·214.9 t/s
2.6 GBS100, Runs great
Unlimited OCR
Q4_K_M·427.0 t/s
3.3 GBS100, Runs great
Gemma 4 E2B IT
Q4_K_M·144.1 t/s
3.5 GBS96, Runs great
North Micro Vision Instruct
Q4_K_M·243.4 t/s
3.7 GBS100, Runs great
NVIDIA Nemotron Labs Dory1.0 4B BF16
Q4_K_M·161.5 t/s
5.7 GBS97, Runs great
FrogNano 4B 2609
Q4_K_M·151.6 t/s
3.4 GBS97, Runs great
Llm Jp 4.1 8B Thinking
Q4_K_M·86.4 t/s
7.6 GBS88, Runs great
D1 2B
Q4_K_M·318.8 t/s
2.3 GBS100, Runs great
WeVisOne VL 4B
Q4_K_M·145.8 t/s
4.7 GBS97, Runs great
WeVisOne VL 2B
Q4_K_M·247.1 t/s
3.6 GBS100, Runs great
Granite Vision 4.1 4B
Q4_K_M·172.2 t/s
4.0 GBS98, Runs great
MiniCPM V 4.6
Q4_K_M·437.3 t/s
1.5 GBS100, Runs great

RTX 3080 specs, power and PSU

Specs

Architecture
Ampere
VRAM
10 GB GDDR6X
Memory bandwidth
760.3 GB/s
FP16 (dense)
59.50 TFLOPS
CUDA cores
8,704
Tensor cores
272
Compute capability
8.6
TDP
320 W
Released
2020-09-17
MSRP
$699

RTX 3080 FAQ

Is the RTX 3080 still worth it in 2026?

At its MSRP of $699 you get 11.1 tok/s per $100 on Qwen3.5 9B, which ranks 1st of the 6 cards compared here; it is also the 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 RTX 3080 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 10 GB, and runs at about 77.7 tok/s.

Can the RTX 3080 run 27–31B models?

No. Qwen3.8 27B needs 17.9 GB at Q4, more than the 10 GB on this card.

How many tokens per second does the RTX 3080 give?

About 77.7 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 3080 need?

Plan a 650 W PSU (2× the 320 W TDP).

Is 10 GB enough for coding and agents?

Yes for coding: Llama 3.1 8B Instruct, the only coding-scored model that fits, runs at about 93 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an RTX 3080?

Rent from $0.14/hr. Renting costs as much as buying after ≈4,900 hours (≈3.3 years at 4 h a day).

Related

Best coding LLMs8 GB VRAM12 GB VRAMGPUs with 10 GB or moreRTX 5080RTX 3080 Ti