Pirate Face
The un-bannable, checksum-verified mirror for open models. Claim your handle before a squatter does.
LiveUpdated 2026-09-19 16:11 UTC
Open & local models
Open & local models scores 59/100 on the AI Radar money board (#6), with 1184 posts and 54.4M views on X over the last 14 days.
It moved up 10 places in the last 24 hours. 88 new tools appeared in this market over the last 7 days, and 459 posts were published about it in the last 72 hours. Jev also rates the room for a newcomer here 2/4.
Top Open & local models tools
atomic.chat
Run AI models locally ->
deepseek-v4.1-flash-exl3-2x-dgx-sparks
DeepSeek v4.1 Flash EXL3 2.9 bpw for 2x DGX Sparks - MiaAI-Lab/DeepSeek-v4.1-Flash-EXL3-2x-DGX-Sparks
bespoke-nimble-9b
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Cactus Compute
It runs on mobiles, wearables, smart home devices, small robots and microcontrollers, with prebuilt engines for macOS, Linux, Windows, Android, iOS, watchOS, tv
orcabonsai-27b-uncensored
Open source
deepseek-v4.1-flash
Original safetensors
k2-horizon-7b-uno
Uno, the speed-up diffusion-augmented version
Intern-S2-397B
🤖 ModelScope
deepseek-v4.1-flash-nvfp4
So Deepseek Dropped Their Weights the Checkpoint however is larger than native weights. So it would be a matter of whether there is a huge quality improvement t
minicpm5-2b
🤗 Hugging Face
laya
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
qwen3.8-flash-next-single-dgx-spark
Qwen3.8-Flash-Next on ONE DGX Spark (TP=1). Contribute to MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark development by creating an account on GitHub.
LM Studio Blog
How to use Splash by Inco AI in LM Studio Bionic
mlx.fast
1 128 GB M5 MacBook, starting at 35 to 36.
OpenEuroLLM
A series of foundation models for transparent AI in Europe
atria-dawn-preview
🤗 Hugging Face
deepseek-v4.1-flash-exl3-dgx-spark-recipe
Contribute to vcruz305/DeepSeek-V4.1-Flash-EXL3-DGX-Spark-recipe development by creating an account on GitHub.
Qwen Studio
Qwen Studio
qwen38-exl3-dflash2
Qwen3.8-27B EXL3 (4.00 bpw) + DFlash2 speculative decoding for ExLlamaV3, validated at 262k context on a 24 GB RTX 3090 - r0b0tlab/qwen38-exl3-dflash2
ternary-bonsai-2-27b-abliterated-gguf
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
qwen38-flashnext-exl3
Contribute to r0b0tlab/qwen38-flashnext-exl3 development by creating an account on GitHub.
ternary-bonsai-2-webgpu-kernels
Go try it out in the Hugging Face space
oui-1-nvfp4
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
FAQ
Is there money in open & local models?
Jev rates buyer spend 1.58/4, search demand 3.3/4, sponsor budgets 1.26/4 and growth 3.05/4. Overall market score: 59/100 (#6 of 19).
What are the newest open & local models tools?
Pirate Face, atomic.chat, deepseek-v4.1-flash-exl3-2x-dgx-sparks, bespoke-nimble-9b, Cactus Compute. Updated 2026-09-19 16:11 UTC.
How is this ranked?
35% money (Jev judgment), 25% traction on X, 20% momentum over 72 hours, 10% new tools, 10% room for newcomers. See methodology.
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. 27k posts from 4.8k X accounts over the last 14 days, 1.2k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 16:11 UTC. Full method.