GPUs & Macs › NVIDIA GTX 1650

NVIDIA GTX 1650: local LLM speed

1 measured result for the NVIDIA GTX 1650.

Measured results

ModelGenerate
tok/s		Prompt
tok/s		QuantBackendTestSource
Llama 3.2 1B Instruct22.1526Q4_K_Mllamafile LocalScore avgLocalScore details

What fits on the NVIDIA GTX 1650

No measured run yet; estimates use spec bandwidth × 0.66 (typical efficiency for NVIDIA). Fit assumes the whole model in GPU memory with ~8K context.

ModelBest quant that fitsFileExpected tok/s
Qwen3.5-4BIQ4_XS2.5 GB33.9 estimate
Qwen3-4BQ4_K_M2.5 GB33.6 estimate
gemma-3-1b-itQ8_01.1 GB70.0 estimate
DeepSeek-R1-0528-Qwen3-8BUD-IQ1_M2.4 GB35.1 estimate
Llama-3.2-3B-InstructQ5_K_M2.3 GB36.2 estimate
Phi-4-mini-instructQ4_K_M2.5 GB33.7 estimate
gemma-3-4b-itQ4_K_M2.5 GB33.7 estimate
Qwen3.5-0.8BQ8_00.8 GB70.0 estimate
Llama-3.2-3B-Instruct-uncensoredQ5_K_M2.6 GB32.4 estimate
Qwen3-0.6BQ8_00.8 GB70.0 estimate
GLM-OCRQ8_01.0 GB70.0 estimate
Qwen3-ASR-1.7BQ8_02.2 GB38.8 estimate
SmolLM3-3BQ4_K_M1.9 GB43.9 estimate
Nanbeige4.2-3BQ4_K_S2.6 GB32.9 estimate

Open in the picker

Similar hardware

NVIDIA GTX 980 · NVIDIA GTX 1050 Ti · NVIDIA GTX 1650 Ti · NVIDIA RTX 2050 · NVIDIA GTX 1050 · NVIDIA GTX 1650 SUPER · NVIDIA Quadro T2000 · NVIDIA GTX 1650 Ti Max-Q