Community Benchmarks
Community Benchmarks is Real benchmark results from oMLX users. Compare performance across Apple Silicon chips and models.. It is ranked #345 on the AI Radar, in Open & local models, first seen 5 days ago and shared in 2 posts (9.9k views).
Community Benchmarks — oMLX Community Benchmarks - Performance Performance Intelligence All chips M1 M2 M3 M4 M5 M6 All variants M1 M1 Max M1 Pro M1 Ultra M2 M2 Max M2 Pro M2 Ultra M3 M3 Max M3 Pro M3 Ultra M4 M4 Max M4 Pro M5 M5 Max M5 Pro M5 Ultra M6 All quants 2bit 3bit 4bit 6bit 8bit others All contexts 1k 4k 8k 16k 32k 64k 128k Apply Hide SpecPrefill results # Chip RAM Model Quant Ctx PP tok/s TG tok/s Note Date 1 M4 (10c) 24 GB Qwen3.5-4B-MLX-4bit 4bit 16k 337.0 30.1 09-21 2 M4 (10c) 24 GB Qwen3.5-4B-MLX-4bit 4bit 8k 344.7 30.8 09-21 3 M4 (10c) 24 GB Qwen3.5-4B-MLX-4bit 4bit 4k 358.0 33.0 09-21 4 M4 (10c) 24 GB Qwen3.5-4B-MLX-4bit…
What people said about Community Benchmarks on X
oMLX community benchmark is back up, home to the largest collection of Apple Silicon MLX benchmark results! https://omlx.ai/benchmarks/performance I got way more submissions than I expected, and my old plan couldn't keep up. So I improved the search logic and upgraded the plan. Depending on how traffic goes I may…
— @jundotkim, 5 days ago · 112 likes · see the post
oMLX 0.7.0.dev4 adds DeepSeek V4.1 CED prefill with up to 79% faster prompt processing on M3 Ultra (opt-in), multi-request Lightning MTP, and plenty of quality-of-life improvements including one-click model settings. https://github.com/jundot/omlx/releases/tag/v0.7.0.dev4 You can now apply model settings with one…
— @jundotkim, 2 days ago · 106 likes · see the post
Alternatives to Community Benchmarks
- Intern-S2-397B — Model Scope
- Ling-3.0-flash-Fin — 🚀 ZDTaichu5.0-9B is now on ModelScope! 🤖
- Pirate Face — The un-bannable, checksum-verified mirror for open models. Claim your handle before a squatter does.
- orcabonsai-27b-uncensored — Open source
- laya — We’re on a journey to advance and democratize artificial intelligence through open source and open science.
- bespoke-nimble-9b — We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Community Benchmarks in numbers
- Rank on the AI Radar: #345 of 1582
- Shared in 2 posts by 1 account: @jundotkim
- 9.9k views on those posts
- First seen 5 days ago, last shared 2 days ago
- Pricing seen by Jev: not stated
- Market: Open & local models
FAQ
What is Community Benchmarks?
Real benchmark results from oMLX users. Compare performance across Apple Silicon chips and models. It was first shared on X 5 days ago and is ranked #345 on the AI Radar.
Is Community Benchmarks free?
Pricing is not stated on the page we read.
Who shared Community Benchmarks?
1 account on X, including @jundotkim, in 2 posts totalling 9.9k views.
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 38k posts from 4.9k X accounts over the last 14 days, 1.6k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-21 07:04 UTC. Full method.