GPUs & Macs › NVIDIA GTX 1080 Ti

NVIDIA GTX 1080 Ti: local LLM speed

Llama 2 7B (Q4_0) generates 71.6 tokens/s on the NVIDIA GTX 1080 Ti (Vulkan), and processes prompts at 648 tokens/s.

Measured results

ModelGenerate
tok/s		Prompt
tok/s		QuantBackendTestSource
Llama 2 7B71.6648Q4_0Vulkan FA ontg128llama.cpp #10879 details
Llama 2 7B67.8586Q4_0Vulkan FA offtg128llama.cpp #10879 details
Llama 2 7B62.51,084Q4_0CUDA FA offtg128llama.cpp #15013 details
Llama 2 7B61.41,138Q4_0CUDA FA ontg128llama.cpp #15013 details
Llama 3.2 1B Instruct1474,098Q4_K_Mllamafile LocalScore avgLocalScore details
Qwen2.5 14B Instruct20.2419Q4_K_Mllamafile LocalScore avgLocalScore details

What fits on the NVIDIA GTX 1080 Ti

Estimates for this chip are scaled from a measured run: Llama 2 7B Q4_0 at 71.6 tok/s (Vulkan, source). Fit assumes the whole model in GPU memory with ~8K context.

ModelBest quant that fitsFileExpected tok/s
Qwen3.8-27BUD-IQ2_XXS7.3 GB37.7 estimate
gemma-4-12B-itQ5_K_M8.4 GB32.5 estimate
Qwen3.5-9BQ6_K7.5 GB36.7 estimate
Qwen3.5-4BQ8_04.5 GB61.0 estimate
MiMo-V2.6-Distill-Qwen-9BQ6_K7.8 GB35.1 estimate
gemma-4-E4B-itQ8_08.2 GB33.4 estimate
Qwen3-4BQ8_04.3 GB63.9 estimate
gemma-4-31B-itUD-IQ2_XXS8.5 GB32.1 estimate
gemma-4-12B-it-qat-q4_0-unquantizedUD-Q4_K_XL6.7 GB40.7 estimate
gemma-3-1b-itQ8_01.1 GB228 estimate
gemma-4-E2B-itQ8_05.0 GB55.1 estimate
DeepSeek-R1-0528-Qwen3-8BQ8_08.7 GB31.4 estimate
Llama-3.2-3B-InstructQ8_03.4 GB80.0 estimate
Phi-4-mini-instructQ8_04.1 GB67.0 estimate

Open in the picker

Similar hardware

NVIDIA RTX 2080 Ti · NVIDIA RTX 3080 · NVIDIA RTX 5070 · NVIDIA TITAN V · NVIDIA TITAN Xp · NVIDIA RTX 3060 · NVIDIA RTX 4070 Ti · NVIDIA RTX 4070 SUPER