BLK · LEADERBOARD

RunLocalAI Score

Every catalog hardware unit ranked by composite score (0–1000): measured tok/s, VRAM fit, ecosystem support, perf-per-watt. 2 of 157 ranks anchored to a measured benchmark — the rest are honestly flagged as extrapolated or estimated.

Methodology: /methodology · Run your own: curl -fsSL runlocalai.co/bench.mjs -o bench.mjs && node bench.mjs

Vendor:All amd apple google intel nvidia other qualcomm

VRAM:All ≤8GB 9–16GB 17–24GB 25–48GB 49–96GB 97GB+

Tier:All S A B C D

Data:All Measured Measured-near Community Extrapolated Estimated

34 units shown · sorted by score

#	Hardware	Tier	Score	Throughput	VRAM-fit	Ecosystem	Efficiency	Data
1	NVIDIA RTX A6000 (Ampere) nvidia · workstation · 48GB	C	477	267	190	200	25	Estimated
2	NVIDIA GeForce RTX 5070 Ti nvidia · high · 16GB	C	477	312	140	200	29	Estimated
3	NVIDIA RTX A5000 nvidia · workstation · 24GB	C	468	267	170	200	32	Estimated
4	NVIDIA A40 nvidia · workstation · 48GB	C	458	242	190	200	22	Estimated
5	NVIDIA GeForce RTX 3080 Ti nvidia · enthusiast · 12GB	C	456	317	110	200	25	Estimated
6	NVIDIA GeForce RTX 3080 12GB nvidia · high · 12GB	C	456	317	110	200	25	Estimated
7	NVIDIA RTX PRO 4000 Blackwell nvidia · workstation · 24GB	C	455	234	170	200	46	Estimated
8	NVIDIA GeForce RTX 4080 Super nvidia · high · 16GB	C	433	256	140	200	22	Estimated
9	NVIDIA GeForce RTX 4080 nvidia · high · 16GB	C	428	250	140	200	22	Estimated
10	NVIDIA GeForce RTX 4070 Ti Super nvidia · high · 16GB	C	418	234	140	200	23	Estimated
11	NVIDIA RTX 5000 Ada Generation nvidia · workstation · 32GB	C	414	200	170	200	22	Estimated
12	NVIDIA RTX 2080 Ti 22GB (China-mod) nvidia · mid · 22GB	C	405	214	140	200	24	Estimated
13	NVIDIA GeForce RTX 4090 Mobile nvidia · enthusiast · 16GB	C	400	200	140	200	32	Estimated
14	NVIDIA GeForce RTX 5070 nvidia · mid · 12GB	C	399	234	110	200	26	Estimated
15	NVIDIA GeForce RTX 3080 10GB nvidia · high · 10GB	C	397	264	80	200	23	Estimated
16	NVIDIA GeForce RTX 5060 Ti 16GB nvidia · mid · 16GB	C	364	156	140	200	24	Estimated
17	NVIDIA GeForce RTX 2080 Ti nvidia · enthusiast · 11GB	C	363	214	80	200	24	Estimated
18	NVIDIA L4 nvidia · workstation · 24GB	C	360	104	170	200	40	Estimated
19	NVIDIA GeForce RTX 3070 Ti nvidia · high · 8GB	C	358	212	80	200	20	Estimated
20	NVIDIA GeForce RTX 4070 nvidia · mid · 12GB	C	356	175	110	200	24	Estimated
21	NVIDIA GeForce RTX 4070 Super nvidia · mid · 12GB	C	355	175	110	200	22	Estimated
22	NVIDIA GeForce RTX 4070 Ti nvidia · high · 12GB	C	351	175	110	200	17	Estimated
23	NVIDIA GeForce RTX 5070 Laptop GPU nvidia · high · 12GB	C	337	134	110	200	37	Estimated
24	NVIDIA GeForce RTX 2080 Super nvidia · high · 8GB	C	330	173	80	200	19	Estimated
25	NVIDIA GeForce GTX 1080 Ti nvidia · high · 11GB	C	327	168	80	200	19	Estimated
26	NVIDIA GeForce RTX 5060 nvidia · entry · 8GB	C	326	156	80	200	29	Estimated
27	NVIDIA GeForce RTX 2060 Super nvidia · mid · 8GB	C	323	156	80	200	25	Estimated
28	NVIDIA GeForce RTX 2070 nvidia · high · 8GB	C	323	156	80	200	25	Estimated
29	NVIDIA GeForce RTX 5060 Ti 8GB nvidia · mid · 8GB	C	322	156	80	200	24	Estimated
30	NVIDIA GeForce RTX 3060 Ti nvidia · high · 8GB	C	321	156	80	200	22	Estimated
31	NVIDIA GeForce RTX 4060 Ti 16GB nvidia · mid · 16GB	C	320	100	140	200	17	Estimated
32	NVIDIA GeForce RTX 3060 12GB nvidia · mid · 12GB	C	319	125	110	200	20	Estimated
33	NVIDIA GeForce RTX 3070 nvidia · mid · 8GB	C	319	156	80	200	20	Estimated
34	NVIDIA GeForce RTX 2070 Super nvidia · high · 8GB	C	319	156	80	200	20	Estimated

BLK · BUY · AMAZON

Shop GPUs & AI hardware on Amazon:GPU categoryRTX 4090RTX 5090Apple M-seriesAI mini-PCs

Amazon search links — we may earn a small commission at no extra cost to you. How we make money.

HOW THE SCORE IS DERIVED

Throughput · 0–500

Steady-state tok/s on a representative 7B/8B Q4 model. Measured from real benchmark rows, or extrapolated from VRAM bandwidth × runtime-stack efficiency.

VRAM-fit · 0–200

How comfortably the rig holds 7B / 32B / 70B class models. Apple unified memory counts; NPU/SoC system RAM counts.

Ecosystem · 0–200

CUDA / MLX / ROCm / Vulkan reach. Real-world friction the operator hits when installing tools.

Efficiency · 0–100

Tok/s per watt. Mobile / NPU class scores well; dense desktop GPUs trade efficiency for absolute throughput.

A confidence multiplier (1.0 measured · 0.85 extrapolated · 0.7 estimated) discounts the headline so we don't pretend to know more than we do. Score is recomputed on every page load against the latest catalog + benchmark data — submit your own run with runlocalai-bench --submit --hardware your-rig to firm up the numbers.

> RunLocalAI Score

RunLocalAI Score