NVIDIA · Turing · 8 GB VRAM tier
Best AI Models for NVIDIA GeForce RTX 2080 SUPER (8 GB)
Runs 7–9B models well; 14B doesn't fit in 8 GB.
Turing 8 GB with Tensor Cores and flash-attention, 496 GB/s; 8 GB caps you at ~7-8B models at Q4.
| # | Model | Speed | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 50.7 t/s | 50.7 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 98.9 t/s | 98.9 tok/s | Q4_K_M · 3.4 GB | |
| 3 | Gemma 2 9B ITGoogle Q4_K_M · 7.3 GB · 52.9 t/s | 52.9 tok/s | Q4_K_M · 7.3 GB | |
| 4 | Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 114 t/s | 114 tok/s | Q4_K_M · 5.0 GB | |
| 5 | Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 237 t/s | 237 tok/s | Q4_K_M · 5.5 GB |
| # | Model | Speed | Reasoning Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 50.7 t/s | 50.7 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 114 t/s | 114 tok/s | Q4_K_M · 5.0 GB | |
| 3 | Qwen2.5 7B InstructAlibaba Q4_K_M · 5.8 GB · 64.6 t/s | 64.6 tok/s | Q4_K_M · 5.8 GB | |
| 4 | Gemma 2 9B ITGoogle Q4_K_M · 7.3 GB · 52.9 t/s | 52.9 tok/s | overall | Q4_K_M · 7.3 GB |
| 5 | Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 237 t/s | 237 tok/s | Q4_K_M · 5.5 GB |
| # | Model | Speed | Long context Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 98.9 t/s | 98.9 tok/s | Q4_K_M · 3.4 GB | |
| 2 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 50.7 t/s | 50.7 tok/s | Q4_K_M · 6.6 GB | |
| 3 | Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 60.6 t/s | 60.6 tok/s | Q4_K_M · 5.6 GB | |
| 4 | Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 237 t/s | 237 tok/s | Q4_K_M · 5.5 GB | |
| 5 | Spark X2.6 4BXHToken Q4_K_M · 4.3 GB · 109 t/s | 109 tok/s | Q4_K_M · 4.3 GB |
| # | Model | Speed ↓ | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 98.9 t/s | 98.9 tok/s | Q4_K_M · 3.4 GB | |
| 2 | Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 237 t/s | 237 tok/s | Q4_K_M · 5.5 GB | |
| 3 | Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 60.6 t/s | 60.6 tok/s | Q4_K_M · 5.6 GB | |
| 4 | Q4_K_M · 6.5 GB · 51.9 t/s | 51.9 tok/s | Q4_K_M · 6.5 GB | |
| 5 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 50.7 t/s | 50.7 tok/s | Q4_K_M · 6.6 GB |
| # | Model | Speed | Vision Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 50.7 t/s | 50.7 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 98.9 t/s | 98.9 tok/s | Q4_K_M · 3.4 GB | |
| 3 | Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 114 t/s | 114 tok/s | Q4_K_M · 5.0 GB | |
| 4 | Q4_K_M · 6.5 GB · 51.9 t/s | 51.9 tok/s | Q4_K_M · 6.5 GB | |
| 5 | LFM3 Omni 2BLiquid AI Q4_K_M · 2.4 GB · 190 t/s | 190 tok/s | Q4_K_M · 2.4 GB |
The RTX 2080 SUPER is one of the fastest 8 GB cards: it runs 7–9B models well. It is power-hungry for its speed, and the best speed per dollar of the cards compared at ≈$240 used. Turing generation, released 2019.
RTX 2080 SUPER strengths and weaknesses
Strengths
- Runs 7–9B models: Qwen3.5 9B at 50.7 tok/s in 6.6 GB
Weaknesses
- 14B models don't fit: Qwen3 14B needs 11.9 GB at Q4 vs 8 GB here
- Little room for context: Qwen3.5 9B uses 6.6 of 8 GB, 1.4 GB left
- No FP8 or NVFP4: Turing 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 2080 SUPER has no new price, so the New view shows it at its MSRP.
| Card | New price | tok/s | tok/s per $100 | tok/s per 100 W |
|---|---|---|---|---|
RTX 50608 GBBest new value | about $460 | 45.8 | 10.0 | 31.6 |
RTX 507012 GB | about $840 | 68.7 | 8.2 | 27.5 |
RTX 5070 Ti16 GB | about $1,170 | 91.6 | 7.8 | 30.5 |
RX 907016 GB | about $800 | 60.4 | 7.6 | 27.5 |
RTX 2080 SUPER8 GB | about $699 | 50.7 | 7.3 | 20.3 |
RTX 40608 GB | about $520 | 27.8 | 5.3 | 24.2 |
| Card | Used price | tok/s | tok/s per $100 | tok/s per 100 W |
|---|---|---|---|---|
RTX 2080 SUPER8 GB | about $240 | 50.7 | 21.1 | 20.3 |
RTX 3060 12GB12 GB | about $329 | 36.8 | 11.2 | 21.6 |
RTX 50608 GBBest used value | about $430 | 45.8 | 10.7 | 31.6 |
RX 907016 GB | about $650 | 60.4 | 9.3 | 27.5 |
RTX 407012 GB | about $575 | 51.5 | 9.0 | 25.8 |
RTX 5070 Ti16 GB | about $1,100 | 91.6 | 8.3 | 30.5 |
RTX 4070 SUPER12 GB | about $675 | 51.5 | 7.6 | 23.4 |
RTX 4070 Ti12 GB | about $722 | 51.5 | 7.1 | 18.1 |
Cost per token: the RTX 2080 SUPER against 8–16 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.
- Card, used price
- $0.30
- Power
- $0.21
Renting costs as much as buying after ≈1,100 hours (≈0.8 years at 4 h a day).
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
Cards people weigh against the RTX 2080 SUPER, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.
Output speed depends on memory bandwidth (496 GB/s). Grey rows need more than 8 GB.
- 9BQwen3.5 9B · Q4_K_M · 6.6 GB50.7tok/s
- 14BQwen3 14B · not at 4-bitneeds 11.9 GB
- 27–31BQwen3.8 27B · not at 4-bitneeds 17.9 GB
- 35B MoEQwen3.6 35B A3B · not at 4-bitneeds 22.0 GB
- 70BLlama 3.3 70B Instruct · not at 4-bitneeds 51.3 GB
- 120B MoEGPT OSS 120B · not at 4-bitneeds 70.8 GB
Estimated speeds: Q4_K_M, 4K context, one user. How we estimate
All 842 models on the RTX 2080 SUPER
Featured onlyBest quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works
| Model | Quant | VRAM | Speed | Context | Score | Fit |
|---|---|---|---|---|---|---|
| Q4_K_M | 7.9 GB | 58.2 t/s | 131K | C44, Usable | ||
Qwen 14BChat | IQ3_S | 7.9 GB | 48.7 t/s | 8K | C44, Usable | |
Qwen2.5 14B InstructChatFunctions | IQ2_XS | 8.0 GB | 62.8 t/s | 33K | C44, Usable | |
Phi 3 Medium 4k InstructChatCode | Q3_K_M | 8.0 GB | 42.8 t/s | 4K | C44, Usable | |
| IQ2_M | 7.8 GB | 66.5 t/s | 131K | B59, Decent | ||
Qwen 14B ChatChat | IQ3_S | 7.9 GB | 48.7 t/s | 8K | C44, Usable | |
DeepSeek v2 LiteChat | Q2_K | 7.5 GB | 185.5 t/s | 164K | B62, Decent | |
| IQ2_S | 7.8 GB | 241.3 t/s | 131K | C44, Usable | ||
Qwen2.5 14B Instruct 1MChatFunctions | IQ2_XS | 8.0 GB | 62.8 t/s | 1010K | C44, Usable | |
Phi 3.5 Mini InstructChatCode | IQ2_S | 7.9 GB | 140.2 t/s | 131K | C44, Usable | |
Phi 3 Medium 128k InstructChatCode | IQ2_XS | 7.8 GB | 65.7 t/s | 131K | C44, Usable | |
Qwen3 14B BaseChat | IQ2_M | 8.0 GB | 57.4 t/s | 33K | C44, Usable | |
Phi 3 Mini 128k InstructChatCode | IQ2_S | 7.9 GB | 140.2 t/s | 131K | C44, Usable | |
DeepSeek Coder v2 Lite BaseChatCode | Q2_K | 7.5 GB | 185.5 t/s | 164K | B62, Decent | |
Qwen2.5 14BChatFunctions | IQ2_XS | 8.0 GB | 62.8 t/s | 131K | C44, Usable |
Specs
- Architecture
- Turing
- VRAM
- 8 GB GDDR6
- Memory bandwidth
- 496.0 GB/s
- FP16 (dense)
- 44.61 TFLOPS
- CUDA cores
- 3,072
- Tensor cores
- 384
- Compute capability
- 7.5
- TDP
- 250 W
- Released
- 2019-07-23
- MSRP
- $699
RTX 2080 SUPER FAQ
Is the RTX 2080 SUPER still worth it in 2026?
At ≈$240 used you get 21.1 tok/s per $100 on Qwen3.5 9B, which ranks 1st of the 8 cards compared here; it is also the 2nd fastest out of 18 cards with 8 GB. It runs 7–9B models well; 14B doesn't fit in 8 GB.
What is the largest model the RTX 2080 SUPER 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 8 GB, and runs at about 50.7 tok/s.
Can the RTX 2080 SUPER run 14B models?
No. Qwen3 14B needs 11.9 GB at Q4, more than the 8 GB on this card.
How many tokens per second does the RTX 2080 SUPER give?
About 50.7 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 3.9 s on a 4K-token prompt.
What power supply does the RTX 2080 SUPER need?
Plan a 500 W PSU (2× the 250 W TDP).
Is renting cheaper than buying an RTX 2080 SUPER?
An RTX 5080 (16 GB) rents from $0.22/hr. Renting costs as much as buying after ≈1,100 hours (≈0.8 years at 4 h a day).