UNIT · NVIDIA · GPU
8 GB VRAMhighReviewed May 2026

NVIDIA GeForce RTX 2070 Super

The Turing refresh that made 8 GB Turing genuinely fast. ~70-90 tok/s on 7B Q4 with ExLlamaV2. Same 8 GB ceiling as the base 2070 but more compute. Strong used-market value in 2026.

Released 2019·~$280 street·448 GB/s memory bandwidth
RUNLOCALAI SCORE
See full leaderboard →
319/ 1000
CC-tier
Estimated
Throughput
156/ 500
VRAM-fit
80/ 200
Ecosystem
200/ 200
Efficiency
20/ 100

Extrapolated from 448 GB/s bandwidth — 53.8 tok/s estimated. No measured benchmarks yet.

Plain-English: Comfortable for 7B chat.

7B chat
Comfortable
14B chat
Doesn't fit
32B chat
Doesn't fit
70B chat
Doesn't fit
Coding agent
Doesn't fit
Vision (≤8B VLM)~
Tight
Long context (32K)
Doesn't fit
Comfortable — fits with headroom
~Tight — works, no slack
Marginal — needs aggressive quant
Doesn't fit usefully

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.

BLK · VERDICT

Our verdict

OP · Fredoline Eruo|VERIFIED MAY 10, 2026
4.8/10

This card is for the operator who wants to run 7B models at high speed on a budget, and is willing to accept the 8 GB VRAM ceiling. It handles 7B Q4 models at ~70-90 tok/s with ExLlamaV2, and can run 13B Q4 models at ~30-40 tok/s with aggressive quantization. The 8 GB limit breaks on 30B+ models, and even 13B models require careful quantization to fit. Pass if you need to run 34B+ models or want headroom for larger context windows. At ~$280 used, it's a strong value for fast 7B inference, but consider an RTX 3060 12 GB if VRAM is the priority.

Why this rating

The RTX 2070 Super offers excellent tok/s for 7B models at a low used price, but the 8 GB VRAM is a hard limit for larger models. It earns a 7.5 for its strong price-to-performance ratio in its niche, but loses points for limited future-proofing.

BLK · OVERVIEW

Overview

The Turing refresh that made 8 GB Turing genuinely fast. ~70-90 tok/s on 7B Q4 with ExLlamaV2. Same 8 GB ceiling as the base 2070 but more compute. Strong used-market value in 2026.

Retailers we'd check:Amazon

Search-fallback links. Editorial hasn't yet curated retailer URLs for this card. Approx. $280.

Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.

BLK · SPECS

Specs

VRAM8 GB
Power draw215 W
Released2019
MSRP$499
Backends
CUDA
Vulkan

Models that fit

Open-weight models small enough to run on NVIDIA GeForce RTX 2070 Super with usable context.

Compare alternatives

Hardware worth comparing

Same VRAM tier and the one step above and below — so you can frame the buying decision against real options.

Frequently asked

What models can NVIDIA GeForce RTX 2070 Super run?

With 8GB VRAM, the NVIDIA GeForce RTX 2070 Super runs 7B models comfortably in Q4 quantization. See the model list below for tested combinations.

Does NVIDIA GeForce RTX 2070 Super support CUDA?

Yes — NVIDIA GeForce RTX 2070 Super is an NVIDIA card with full CUDA support, the most mature local-AI backend. llama.cpp, Ollama, vLLM, and ExLlamaV2 all run natively.

How much does NVIDIA GeForce RTX 2070 Super cost?

Current street price for NVIDIA GeForce RTX 2070 Super is around $280 (MSRP $499). Prices vary by region and supply.

Where next?

Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify hardware specifications.