NVIDIA
Operating mode
Use this to bias workload recommendations toward responsiveness, background autonomy, lighter serving, or multi-GPU scale-out.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
The RTX 2080 Ti 11GB was NVIDIA's 2018 flagship and remains capable for local AI inference. At 11 GB VRAM, it can run 7B models at FP16 and 13B models at Q4 — more headroom than any 8 GB card. Its 616 GB/s bandwidth is solid, but the 2nd-gen Tensor Cores are notably less efficient than Ampere or Ada equivalents for inference. Studies have shown it achieves similar tokens/sec to an RTX 4060 Ti despite higher raw specs — a sign of software optimizations favoring newer architectures. Still excellent value used for VRAM at this price tier.
Beyond LLMs
What AI tasks this GPU can handle — from text generation to image and video creation.
| Capability | Status | Representative Model |
|---|---|---|
| LLM Chat (7B) | Runs natively | Llama 3.1 8B Q4 |
| LLM Coding (30B) | Won’t fit | Qwen 3 30B Q4 |
| LLM Large (70B) | Won’t fit | Llama 3.1 70B Q4 |
| Image Gen (SDXL) | Runs natively | SDXL 1.0 FP16 |
| Image Gen (Flux) | Won't fit | Flux.1 Dev FP16 |
| Image Gen (SD 3.5) | Won't fit | SD 3.5 Large FP16 |
| Video Short (25f) | Runs with offload | LTX Video 2B |
| Video Long (100f) | Won't fit | Wan Video 14B |
Architecture
Turing is NVIDIA's first-generation RTX architecture, introducing dedicated RT and Tensor Cores to consumer GPUs for the first time. Built on TSMC's 12nm FinFET process.
AI Relevance
The first consumer architecture with Tensor Cores, enabling meaningful acceleration for INT8 and FP16 inference. However, limited VRAM (typically 6-11 GB) restricts modern LLM model sizes.
Conselho de compra
Utilizável para IA local com limitações
Pode rodar 9 de 50 modelos principais, principalmente os menores. Modelos maiores precisam de quantização forte ou não cabem.
11.0 GB
VRAM
$999
Preço sugerido
$91/GB
Custo por GB de VRAM
Melhores modelos para esta GPU
What will limit you first
This setup is broadly balanced for this model.
Older PCIe generation
PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.
Best upgrade itinerary
Unlocks 3 additional models that do not fit on the current setup.
Quer mais margem? RTX 3060 12GB (12.0 GB VRAM) é o próximo passo.
Chat
SThis model is a direct match for chat. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Coding
SThis model is a direct match for coding. It belongs to a current frontier family for local AI. It should run, but memory headroom will be limited. Known channels: huggingface, ollama, lm-studio.
Agentic Coding
AThis model is still usable for agentic-coding, but it is not the most specialized pick. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama.
Reasoning
SThis model is a direct match for reasoning. It belongs to a current frontier family for local AI. It should run, but memory headroom will be limited. Known channels: huggingface, ollama, lm-studio.
RAG
AThis model is still usable for rag, but it is not the most specialized pick. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama.
Qwen 3.5 9BQwen 3 8BQwen 3.5 4BNemotron Nano 8B
Phi-4 Mini Reasoning 4B
Jina Embeddings v3
BGE M3Qwen 3 14B
Ministral 3 14BQwen3-Coder 30B A3B InstructQwen 3.5 397B A17BDevstral 2 123B Instruct
Kimi K2.5Kimi K2.6
DeepSeek V4 ProQwen 3.5 27BQwen 3.6 27BQwen 3.5 122B A10BQwen3-VL 30B A3B InstructQwen 3.6 35B A3BDeepSeek V4 FlashQwen 3.5 35B A3BMagistral Small 2507Devstral Small 2 24B InstructQwen 3 32BQwen 3 30B A3BMistral Small 4 119B
Command A 111BQwen 2.5 VL 72B
GPT-OSS 120BNemotron 3 Nano 30BQwen3-Coder-NextPhi-4-reasoning-plus 14BDevstral Small 1.1
GLM-5.1Pixtral Large 124BGLM-5DeepSeek V3.2GPT-OSS 20BQwen 3 235B A22BQwen3-Coder 480B A35B InstructNemotron Cascade 2 30B A3B
Gemma 4 31BLeanstral 119B A6BDeepSeek Coder V2 236BDeepSeek R1 671BDeepSeek V3.1 671B
EXAONE 4.0 32BGemma 4 26B A4BQuase ao alcance
Modelos de alta qualidade que precisam de um pouco mais de memória
Image & Video Generation
23 of 52 models can generate images or video on your RTX 2080 Ti 11GB
| Model | Max Resolution | Gen Time | Grade |
|---|---|---|---|
| SD TurboImage | 512×512 | ~1.8s | S |
| Stable Diffusion 1.5Image | 512×768 | ~3.6s | S |
| Realistic Vision v5.1Image | 512×768 | ~3.6s | S |
| DreamShaper 8Image | 512×768 | ~3.6s | S |
| LCM DreamShaper v7Image | 512×768 | ~1.1s | S |
| PixArt-SigmaImage | 256×256 | ~14.2s | S |
| FramePack I2VVideo | 256×256 | ~26.1s/frame | S |
| SDXL TurboImage | 512×512 | ~1.8s | S |
| SDXL LightningImage | 1024×1024 | ~5.3s | S |
| Stable Diffusion XL 1.0Image | 1024×1024 | ~14.2s | S |
| Playground v2.5Image | 1024×1024 | ~21.3s | S |
| RealVisXL v5.0Image | 1024×1024 | ~16s | S |
| DreamShaper XLImage | 1024×1024 | ~16s | S |
| Juggernaut XL v9Image | 1024×1024 | ~16s | S |
| Animagine XL 3.1Image | 1024×1024 | ~16s | S |
| Pony Diffusion V6 XLImage | 1024×1024 | ~16s | S |
| Animagine XL 4.0Image | 1024×1024 | ~16s | S |
| Illustrious XLImage | 1024×1024 | ~16s | S |
| Wan Video 2.1 1.3BVideo | 256×256 | ~10.4s/frame | A |
| Stable Diffusion 3.5 MediumImage | 256×256 | ~24.9s | A |
| Flux.2 Klein 4BImage | 256×256 | ~4.3s | A |
| LTX Video 2BVideo | 256×256 | ~12.3s/frame | B |
| KolorsImage | 256×256 | ~28.4s | D |
| Stable CascadeImage | 1024×1024 | ~35.5s | F |
| AuraFlow v0.3Image | 256×256 | ~1m 4s | F |
| Stable Diffusion 3.5 LargeImage | 256×256 | ~1m 18s | F |
| Stable Diffusion 3.5 Large TurboImage | 256×256 | ~14.2s | F |
| CogVideoX 2BVideo | 256×256 | ~12.3s/frame | F |
| HunyuanVideoVideo | 256×256 | ~26.1s/frame | F |
| ChromaImage | 256×256 | ~14.2s | F |
| Z-Image TurboImage | 256×256 | ~14.7s | F |
| Flux.1 DevImage | 256×256 | ~1m 4s | F |
| Flux.1 SchnellImage | 256×256 | ~12.4s | F |
| LTX Video 13BVideo | 256×256 | ~26.1s/frame | F |
| Flux.1 Kontext DevImage | 256×256 | ~1m 11s | F |
| AnimateDiff v1.5.3Video | 512×768 | ~6.5s/frame | F |
| Cosmos Diffusion 7BVideo | 256×256 | ~20.4s/frame | F |
| CogVideoX 5BVideo | 256×256 | ~17.8s/frame | F |
| Wan2.2 TI2V 5BVideo | 256×256 | ~17.8s/frame | F |
| Flux.2 Klein 9BImage | 256×256 | ~7.1s | F |
| Flux.1 Fill DevImage | 256×256 | ~1m 0s | F |
| Mochi 1 PreviewVideo | 256×256 | ~23.5s/frame | F |
| HunyuanVideo 1.5Video | 256×256 | ~21.8s/frame | F |
| Helios 14BVideo | 256×256 | ~26.9s/frame | F |
| SkyReels V2 14BVideo | 256×256 | ~26.9s/frame | F |
| Wan Video 2.1 14BVideo | 256×256 | ~26.9s/frame | F |
| Wan Video 2.2 14BVideo | 256×256 | ~26.9s/frame | F |
| Qwen ImageImage | 256×256 | ~23.9s | F |
| Qwen Image EditImage | 256×256 | ~23.9s | F |
| Flux.2 DevImage | 256×256 | ~11m 12s | F |
| MAGI-1Video | 256×256 | ~33.3s/frame | F |
| HunyuanImage 3.0Image | 256×256 | ~42.1s | F |
Image models estimated at 1024×1024 (28 steps, FP16). Video models estimated at 768×512 (25 frames, 30 steps, FP16). Actual performance varies with runtime and system load.
Upgrade paths
See what you unlock with more powerful hardware
Opções de upgrade
RTX 3060 12GBPróximo passoUnlocks 3 additional models that do not fit on the current setup.
~$329 MSRP
RTX 4070 12GBUpgrade NVIDIAUnlocks 3 additional models that do not fit on the current setup.
~$599 MSRP
Intel Arc Pro B60 24GBMelhor custo-benefícioUnlocks 76 additional models that do not fit on the current setup.
~$599 MSRP
Unlocks 121 additional models that do not fit on the current setup.
Lifts average decode speed across fitting models by about 148%.
~$8,000 MSRP
RTX 2080 Ti 11GB (11 GB VRAM) can run these top models: Qwen 3.5 9B (score: 95/100), Qwen 3 8B (score: 94/100), Qwen 3.5 4B (score: 93/100). See the full compatibility list above.
RTX 2080 Ti 11GB has 11 GB of VRAM available for AI model inference. This determines which models and quantization levels you can run locally.
Yes, RTX 2080 Ti 11GB is excellent for running LLMs locally with top compatibility scores above 80/100.
For coding on RTX 2080 Ti 11GB, we recommend Qwen 3.5 9B. It achieves 78.4 tokens per second with 26K context window. This model is a direct match for coding. It belongs to a current frontier family for local AI. It should run, but memory headroom will be limited. Known channels: huggingface, ollama, lm-studio.
There are 4 upgrade path(s) from RTX 2080 Ti 11GB: RTX 3060 12GB, RTX 4070 12GB. Upgrading would unlock larger models and faster inference speeds.
Flux.1 Dev requires around 24 GB of usable memory at FP16. With 11 GB, RTX 2080 Ti 11GB cannot run Flux natively. Consider quantized GGUF variants or the smaller Schnell model with aggressive offloading.
RTX 2080 Ti 11GB (11 GB VRAM) can handle various AI generation tasks beyond LLMs. For image generation, SDXL and Stable Diffusion 3.5 run well. For video, video generation is limited by available memory. Check the AI Capability Matrix above for detailed compatibility.
RTX 2080 Ti 11GB can handle basic AI image generation with SDXL and SD 1.5. With 11 GB of usable memory, larger models like Flux will need quantization or offloading. Best suited for standard resolution (512-1024px) generation.
Qwen 3.5 27B does not fit on RTX 2080 Ti 11GB with 11 GB. However, Qwen 3.5 9B at Q4 (5.5 GB) or Q5 (6.5 GB) runs well on your GPU. The 4B variant fits at Q8 for near-lossless quality.
With 11 GB on RTX 2080 Ti 11GB, use Q4_K_M for 8B models and Q4_K_M with tight context for 14B models. Q5_K_M is a good middle ground when the model fits. For the best quality-to-size ratio, Q4_K_M is the most popular choice.
On RTX 2080 Ti 11GB, capacity is usually the first gate: if the model does not fit, bandwidth does not matter. But once a model fits, memory bandwidth is what largely determines tokens per second. In practice, you want enough memory to fit the model plus headroom, then as much bandwidth as your budget allows.
Compare with similar
