alkinun

AtAndDev

·

https://baretile.com/

alkinun

AI & ML interests

decentralize

Recent Activity

liked a model about 6 hours ago

moonshotai/Kimi-K3.1-Pro

liked a model about 10 hours ago

alpindale/lexnet

reacted to savoji's post with 🚀 about 10 hours ago

LFM3-1.7B from @LiquidAI is the #1 trending model on the Hub (855 likes). It takes about 1 GB at 4-bit, and in the vendor's own table it beats Qwen3.5 9B on several agent benchmarks. The edge decision moves from "are small models good enough?" to "which architecture and runtime do I pick?" One 1.7B backbone now ships as text, VL-2B and Omni-2B checkpoints with open weights. 🔍 Architecture and speed • 24 layers: 13 gated convolutions (kernel 7), 5 sliding-window attention (256 tokens), 6 global attention, chosen by a search over 14 device and backend targets • llama.cpp Q4_0 at 32K context: 1.8 to 2.8x faster decode than the next model on 3 devices • Full 128K context needs 1.6 GB of KV cache vs 4.3 GB (Qwen3.5-4B) and 5.6 GB (MiniCPM5-2B) 📊 Liquid AI's table • Average 49.37 vs 47.24 for Qwen3.5 9B; BFCL V4 71.39 vs 62.83; WebArena-Verified 38.42 vs 28.94 • Qwen3.5 9B still wins SWE-bench Verified (51.20 vs 43.40) and GPQA Diamond (80.40 vs 63.37) 🛠️ Try it • GGUF (174,177 downloads last month) runs in llama.cpp b11960+, Ollama and LM Studio; MLX 4-bit for Apple Silicon; SGLang for GPUs • llama-cli -hf LiquidAI/LFM3-1.7B-GGUF:Q4_K_M --temp 0.1 --top-k 50 --repeat-penalty 1.1 • Omni-2B runs in the browser in the WebGPU Space ⚠️ Caveats • The numbers are vendor-reported, with no independent reproduction in the sources. The card says it is not for agentic coding or complex multi-step reasoning. https://huggingface.co/LiquidAI/LFM3-1.7B https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF https://huggingface.co/LiquidAI/LFM3-Omni-2B https://huggingface.co/spaces/LiquidAI/LFM3-Omni-2B-WebGPU Has a sub-2B model held up on your own tool-calling harness? • Medium: https://medium.com/p/05507e4e81c8 • Substack: https://sajjadpakdamansavoji.substack.com/p/liquid-ais-lfm3-turns-the-small-model • X: https://x.com/saavoji/status/2117307514177941672 • Website: https://sajjadpsavoji.github.io/blog#liquid-ais-lfm3

View all activity

Organizations

AtAndDev 's models 20

AtAndDev/UVOX-Magpie-xLAM-Supernova-8B

0.7B • Updated Oct 30, 2025 • 13

AtAndDev/Magpie-xLAM-Supernova-8B

Text Generation • 8B • Updated Oct 30, 2025 • 16 • 1

AtAndDev/UVOX-xLAM-2-8B

0.7B • Updated Oct 25, 2025 • 12

AtAndDev/UVOX-96k-Llama-32-3B-Instruct

Audio-Text-to-Text • 49.6M • Updated Oct 25, 2025 • 22

AtAndDev/UVOX-40k-Llama-32-3B-Instruct

Audio-Text-to-Text • 49.6M • Updated Oct 23, 2025 • 24

AtAndDev/UVOX-12k-Llama-32-3B-Instruct

Audio-Text-to-Text • 49.6M • Updated Oct 23, 2025 • 20

AtAndDev/UVOX-95k-Llama-3.2-1B-Instruct

Audio-Text-to-Text • 46.1M • Updated Oct 21, 2025 • 24

AtAndDev/UVOX-90k-Llama-3.2-1B-Instruct

Audio-Text-to-Text • 46.1M • Updated Oct 21, 2025 • 33

AtAndDev/UVOX-50k-Llama-3.2-1B-Instruct

Audio-Text-to-Text • 46.1M • Updated Oct 21, 2025 • 27

AtAndDev/UVOX-5k-Llama-3.2-1B-Instruct

Audio-Text-to-Text • 46.1M • Updated Oct 20, 2025 • 28

AtAndDev/UVOX-Qwen3-4B-Instruct-2507

Audio-Text-to-Text • 47.2M • Updated Oct 20, 2025 • 21

AtAndDev/SelfCoder-v0.0

Text Generation • 4B • Updated May 19, 2025 • 16

AtAndDev/lora_model

Updated Apr 5, 2025

AtAndDev/marco-qwq-7B

Text Generation • 8B • Updated Dec 8, 2024 • 25

AtAndDev/Ogno-Monarch-Neurotic-9B-Passthrough

Text Generation • 9B • Updated Mar 1, 2024 • 63

AtAndDev/Ogno-Monarch-Neurotic-7B-Dare-Ties

Text Generation • 7B • Updated Mar 1, 2024 • 59

AtAndDev/Marcoro14-7B-Slerp

Text Generation • 7B • Updated Mar 1, 2024 • 9

AtAndDev/CapybaraMarcoroni-7B

Text Generation • 7B • Updated Jan 7, 2024 • 60

AtAndDev/ShortKing-3b-v0.2

Text Generation • 3B • Updated Oct 2, 2023 • 66 • 2

AtAndDev/ShortKing-1.4b-v0.1

Text Generation • 1B • Updated Sep 29, 2023 • 40 • 2