8.3B (dense)
Apache 2.0
First model from Essential AI (founded by Ashish Vaswani); exceptional code generation and agentic capabilities; leads 8B class on SWE-bench Verified (20.8%, beating Gemini 2.0 Flash and Qwen2.5-Coder 32B); SOTA tool use on BFCL; strong math (AIME) and STEM (GPQA); 32K context with YaRN extension; trained on 8.4T tokens using Muon optimizer instead of AdamW; architecture similar to Gemma 3 with global self-attention; Base and Instruct variants designed for community specialization; excels in multi-language code generation and FIM tasks.