Weekly archive
New local AI models, May 18–24, 2026
4 open models released May 18–24, 2026, with how each runs on five reference devices.
-
92Excellent quality (A+) · 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking.
Runs on
-
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.
Runs on
-
82Great quality (A) · 82 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.
Runs on
- Phone iPhone 17 Pro Runs welltight fit 10.5 tok/s
- Laptop MacBook Pro M5 Pro 64 GB Runs well 27.1 tok/s
- Gaming PC RTX 5090 32 GB Runs great 43.2 tok/s
- Edge board Jetson Orin NX 16 GB Runs well 10.5 tok/s
- Workstation Mac Studio 512 GB Runs great 50.2 tok/s
-
36Basic quality (D) · 36 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.
Runs on
- Phone iPhone 17 Pro Runs great 45.7 tok/s
- Laptop MacBook Pro M5 Pro 64 GB Runs great 70.5 tok/s
- Gaming PC RTX 5090 32 GB Runs great 478 tok/s
- Edge board Jetson Orin NX 16 GB Runs great 59.3 tok/s
- Workstation Mac Studio 512 GB Runs great 188 tok/s
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.