llmrun

NVIDIA · Ampere · 12 GB VRAM tier

Best AI Models for NVIDIA GeForce RTX 3080 12GB

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

12 GB GDDR6X912 GB/s61 TFLOPS FP16350 WAmpere 2022
Buyabout $799

Ampere 12 GB at 912 GB/s: fits ~13-14B at Q4 and decodes fast, but draws 350 W. A different, faster chip config than the 10 GB RTX 3080.

What LLMs can the RTX 3080 12GB run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 93.2 t/s
93.2 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 182 t/s
182 tok/sQ4_K_M · 3.4 GB
3
Gemma 3 12B ITGoogle
Q4_K_M · 10.9 GB · 73.7 t/s
73.7 tok/sQ4_K_M · 10.9 GB
4
Gemma 2 9B ITGoogle
Q4_K_M · 7.3 GB · 97.2 t/s
97.2 tok/sQ4_K_M · 7.3 GB
5
MiMo V2.6 Distill Qwen 9BXiaomi
Q4_K_M · 6.5 GB · 95.5 t/s
95.5 tok/sQ4_K_M · 6.5 GB
See all 278 models on the RTX 3080 12GB →Best quant that fits · speeds estimated on this card

About the RTX 3080 12GB

The RTX 3080 12GB is the quickest card in the 12 GB class: it runs 7–9B models well. It is among the most efficient per watt in its tier, and the best speed per dollar of the cards compared at ≈$513 used. Ampere generation, released 2022.

Speed in 12 GB#1912 GB/s
Per watt#20.27 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#118.2 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationAmperereleased 2022

RTX 3080 12GB strengths and weaknesses

Strengths

  • Runs 14B models: Qwen3 14B at 62.4 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
  • 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 12GB has no new price, so the New view shows it at its MSRP.

CardNew pricetok/stok/s per $100tok/s per 100 W
RTX 3080 12GB12 GB
about $79993.2
11.7
26.6
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 9060 XT 16GB16 GB
about $59030.2
5.1
18.9
RTX 508016 GB
about $2,39498.1
4.1
27.3
RTX 509032 GB
about $5,502183
3.3
31.8
CardUsed pricetok/stok/s per $100tok/s per 100 W
RTX 3080 12GB12 GB
about $51393.2
18.2
26.6
RTX 3060 12GB12 GB
about $32936.8
11.2
21.6
RTX 50608 GBBest used value
about $43045.8
10.7
31.6
RTX 507012 GB
about $73668.7
9.3
27.5
RTX 5070 Ti16 GB
about $1,10091.6
8.3
30.5
RTX 3090 Ti24 GB
about $1,600103
6.4
22.9
RX 9060 XT 16GB16 GB
about $55030.2
5.5
18.9
RTX 509032 GB
about $7,000183
2.6
31.8

Cost per token: the RTX 3080 12GB 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 GB12 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 5090 · 32 GB · ≈$7,000 · $2.56 per 1M tokensRTX 5070 · 12 GB · ≈$736 · $0.83 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRTX 3090 Ti · 24 GB · ≈$1,600 · $1.17 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$550 · $1.38 per 1M tokensRTX 5060 · 8 GB · ≈$430 · $0.73 per 1M tokensRTX 5080 · 16 GB · ≈$2,000 · $1.45 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 tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRTX 4060 · 8 GB · ≈$335 · $0.94 per 1M tokensRTX 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 4060 Ti 8GB · 8 GB · ≈$400 · $1.09 per 1M tokensRTX 3080 12GB · 12 GB · ≈$513 · $0.51 per 1M tokensRTX 3080 12GBRTX 5090RTX 3090 TiRTX 5080RTX 5070 TiRTX 4080
$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 5090 · 32 GB · ≈$7,000 · $2.56 per 1M tokensRTX 5070 · 12 GB · ≈$736 · $0.83 per 1M tokensRTX 3060 12GB · 12 GB · ≈$329 · $0.76 per 1M tokensRTX 3090 Ti · 24 GB · ≈$1,600 · $1.17 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$550 · $1.38 per 1M tokensRTX 5060 · 8 GB · ≈$430 · $0.73 per 1M tokensRTX 5080 · 16 GB · ≈$2,000 · $1.45 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 tokensRX 9070 · 16 GB · ≈$650 · $0.83 per 1M tokensRTX 4060 · 8 GB · ≈$335 · $0.94 per 1M tokensRTX 4070 · 12 GB · ≈$575 · $0.87 per 1M tokensRTX 4060 Ti 8GB · 8 GB · ≈$400 · $1.09 per 1M tokensRTX 3080 12GB · 12 GB · ≈$513 · $0.51 per 1M tokensRTX 3080 12GBRTX 5090RTX 3090 Ti
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$0.51
Card, used price
$0.35
Power
$0.16
3-year cost of ownership$743
Or rent an RTX 5080 in the cloud$0.22/hr

Renting costs as much as buying after ≈2,400 hours (≈1.6 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 12GB, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

12 GB more and 11% fasterRTX 3090 TiNVIDIA · 24 GB · Ampere
▲ 11% speed▲ 12 GBabout $1,999rent $0.27/hr
20 GB more and nearly 2× the speedRTX 5090NVIDIA · 32 GB · Blackwell
▲ 97% speed▲ 20 GBabout $5,502rent $0.51/hr
61% slower, same 12 GBRTX 3060 12GBNVIDIA · 12 GB · Ampere
▼ 60% speedsame 12 GBabout $329rent $0.07/hr
Same 12 GB, new with warrantyRTX 5070NVIDIA · 12 GB · Blackwell
▼ 26% speedsame 12 GBabout $840rent $0.15/hr
4 GB more, but slowerRX 9060 XT 16GBAMD · 16 GB · RDNA 4
▼ 68% speed▲ 4 GBabout $590
About as fast, same 12 GBRTX 3080 TiNVIDIA · 12 GB · Ampere
same speedsame 12 GBabout $1,199rent $0.13/hr

RTX 3080 12GB tokens per second by model size

Output speed depends on memory bandwidth (912 GB/s). Grey rows need more than 12 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    93.2tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB tight · Q4_K_M
    62.4tok/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 977 models on the RTX 3080 12GB

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 12GB — ranked by Score, recency and Fit
ModelVRAMScoreFit
LFM2.5 350M
Q4_K_M·1022.1 t/s
0.8 GBS100, Runs great
DeepSeek R1 Distill Qwen 1.5B
Q4_K_M·417.5 t/s
1.8 GBS100, Runs great
Qianfan OCR
Q4_K_M·178.0 t/s
4.7 GBS99, Runs great
Phi 4
Q4_K_S·66.1 t/s
11.9 GBC44, Usable
Arete 27B
IQ3_XXS·52.2 t/s
11.8 GBC44, Usable
LightOnOCR 3 1B
Q4_K_M·581.2 t/s
1.8 GBS100, Runs great
Dots.mocr
Q4_K_M·271.9 t/s
2.6 GBS100, Runs great
Turkish Gemma 9B T1
Q4_K_M·91.8 t/s
7.7 GBS90, Runs great
InternVL2 2B
Q4_K_M·406.0 t/s
3.3 GBS100, Runs great
Mathstral 7B v0.1
Q4_K_M·120.5 t/s
6.8 GBS95, Runs great
Schematron 8B
Q4_K_M·110.4 t/s
7.3 GBS93, Runs great
Jina Ocr V1
Q4_K_M·475.6 t/s
3.3 GBS100, Runs great
GPT OSS 20B NPU2
Q3_K_M·242.8 t/s
10.3 GBS92, Runs great
Tiny Random Gemma2ForCausalLM
Q4_K_M·1912.3 t/s
0.3 GBS100, Runs great
Eurus 2 7B PRIME
Q4_K_M·118.8 t/s
5.1 GBS95, Runs great

RTX 3080 12GB specs, power and PSU

Specs

Architecture
Ampere
VRAM
12 GB GDDR6X
Memory bandwidth
912.0 GB/s
FP16 (dense)
61.28 TFLOPS
CUDA cores
8,960
Tensor cores
280
Compute capability
8.6
TDP
350 W
Released
2022-01-11
MSRP
$799

RTX 3080 12GB FAQ

Is the RTX 3080 12GB still worth it in 2026?

At ≈$513 used you get 18.2 tok/s per $100 on Qwen3.5 9B, which ranks 1st of the 8 cards compared here; it is also the fastest out of 8 cards with 12 GB. It runs 7–9B models well; 14B only just fits.

What is the largest model the RTX 3080 12GB 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 191 tok/s.

Can the RTX 3080 12GB 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 3080 12GB 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 3080 12GB give?

About 93.2 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 2.9 s on a 4K-token prompt.

What power supply does the RTX 3080 12GB need?

Plan a 700 W PSU (2× the 350 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 112 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an RTX 3080 12GB?

An RTX 5080 (16 GB) rents from $0.22/hr. Renting costs as much as buying after ≈2,400 hours (≈1.6 years at 4 h a day).

Related

Best coding LLMs12 GB VRAM16 GB VRAMGPUs with 12 GB or moreRTX 4090RX 7900 XTX