Tagged: open-source-llm

15 articles

Latest

IBM's Granite 4.2 packs 512K context and agentic tool use into Apache 2.0 models

IBM's Granite 4.2 goes all-in on chain-of-thought reasoning with 128K context

StarCoder's 15B model beats OpenAI's original Codex on key coding benchmarks

Vicuna-13B runs on one AMD GPU for $300 — here's the ROCm setup

Falcon-40B Lands on Hugging Face: 90GB of GPU for a Free GPT-3 Rival

Falcon 180B: The 180B-parameter open model beating GPT-3.5 on MMLU

Falcon-Arabic 7B beats models 4x its size on Arabic benchmarks

Gemma 3's 27B model punches into top 10 of Chatbot Arena, beating Gemini 1.5 Pro

Falcon Mamba: The First 7B Model to Ditch Attention Without Performance Loss

Llama 3.1 lands: 405B parameters, 128K context, and a license that lets you distill

Google's Gemma 2 drops: A 27B open model that beats Llama 3 70B

Falcon 2 hits 11B parameters, matching Falcon-40B at a quarter of the size

Meta drops Llama 3: 15 trillion training tokens, 8K context, and a new 128K tokenizer

Google's Gemma 7B matches Mistral on benchmarks as open model race tightens

Fine-tune with Mixtral: $2.70 beats GPT-4's $3,061 price tag