NVIDIA GeForce GTX 1650 Super
No editorial image yet — generic vendor mark shown. Credentials in spec table below.
Turing entry refresh with GDDR6. 4 GB VRAM is below the practical AI floor — 1-3B Q4 only. No Tensor cores. Common in pre-built office PCs from 2019-2021 that are now hitting the AI question. The honest answer for this card is usually 'try CPU offload or upgrade.'
Affiliate disclosure: as an Amazon Associate and partner of other retailers, we earn from qualifying purchases. The verdict on this page is our editorial opinion; affiliate links never influence what we recommend.
Sub-scores sum to 315 / 1000. Headline = 315 × 0.70 (Estimated-confidence discount) = 221. This is an algorithmic performance-tier score — distinct from, and often lower than, the editorial “Our verdict” below, which weighs value and real-world fit (especially for hardware we haven’t measured yet). How scoring works →
Extrapolated from 192 GB/s bandwidth — 23.0 tok/s estimated. No measured benchmarks yet.
Plain-English: Doesn't fit modern chat models usefully — vision models won't fit.
Verdicts extrapolated from catalog VRAM + bandwidth + ecosystem flags. Hover any chip for the rationale. Want measured numbers? Submit your own run with runlocalai-bench --submit.
This card is for the operator who already owns a GTX 1650 Super in a pre-built office PC and wants to know if it can run any local AI at all. The answer is marginal: 1-3B Q4 models run at 20-35 tok/s, usable for simple chat or code completion. 7B models are out of reach due to 4 GB VRAM, though CPU offload with llama.cpp can squeeze a 7B Q2 at ~2-4 tok/s — painful but technically possible. What breaks: anything above 3B parameters, any model requiring 4-bit or higher quantization, and any workload needing Tensor Cores (none present). Operators should pass if they plan to run 7B+ models, want reasonable inference speed, or can spend a bit more on a used RTX 3060 12 GB. At ~$140 used, this card is only worth it if the budget is absolutely zero and the task is strictly 1-3B models.
›Why this rating
The 4 GB VRAM is below the practical floor for most local AI workloads, and the lack of Tensor Cores limits performance even on small models. It barely qualifies for 1-3B Q4 models, but the value is poor compared to similarly priced used options with more VRAM.
Overview
Turing entry refresh with GDDR6. 4 GB VRAM is below the practical AI floor — 1-3B Q4 only. No Tensor cores. Common in pre-built office PCs from 2019-2021 that are now hitting the AI question. The honest answer for this card is usually 'try CPU offload or upgrade.'
Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.
Specs
| VRAM | 4 GB |
| Power draw (peak) | 100 W |
| Released | 2019 |
| MSRP | $159 |
| Backends | CUDA Vulkan |
Models that fit
Open-weight models small enough to run on NVIDIA GeForce GTX 1650 Super with usable context.
Frequently asked
What models can NVIDIA GeForce GTX 1650 Super run?
Does NVIDIA GeForce GTX 1650 Super support CUDA?
How much does NVIDIA GeForce GTX 1650 Super cost?
Where next?
Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify hardware specifications.