Weekly archive

New local AI models, Oct 14–20, 2024

2 open models released Oct 14–20, 2024, with how each runs on five reference devices.

Allegro (2.8B)

Rhymes AI · Video · 2.8B · released Oct 16

52Fair quality (C) · 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking.

Runs on
Memory, every quant and how to run it Check your own device
Ministral 8B (2410)

Mistral AI · LLM · 8.0B · released Oct 15

46Fair quality (C) · 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.

Runs on
- Phone iPhone 17 Pro Runs welltight fit 12.1 tok/s
- Laptop MacBook Pro M5 Pro 64 GB Runs great 40.5 tok/s
- Gaming PC RTX 5090 32 GB Runs great 64.4 tok/s
- Edge board Jetson Orin NX 16 GB Runs well 15.7 tok/s
- Workstation Mac Studio 512 GB Runs great 55.0 tok/s
Memory, every quant and how to run it Check your own device

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.

Allegro (2.8B)

Ministral 8B (2410)