StarCoder2 and The Stack v2

BigCode announced the release of the StarCoder2 family of open code LLMs in three sizes (3B, 7B, 15B) and The Stack v2 dataset. The models use a 16,384-token context window and are licensed under BigCode OpenRAIL-M v1.

From the source
BigCode is releasing StarCoder2, the next generation of transparently trained open code LLMs.
huggingface.co

Visit source ↗huggingface.co

Announced: 28 Feb 2024 00:00 UTC
Detected: 03 Aug 2026 21:12 UTC (+887d after)
License: BigCode OpenRAIL-M v1
Context window

Other stories

Google Research · 31 AugTimesFM-3: A zero-shot foundation model for multivariate forecastingNew model

Hugging Face · 28 AugThe Open ASR Leaderboard Adds Its First Global South LanguageCapability change

Perplexity · 27 AugDeepSeek V4 Flash 0731New model

Tencent · 27 AugIntroducing Hy4 previewNew model

16,384 tokens

Amazon · 31 AugConnect an AgentCore Runtime hosted MCP server to Amazon QuickCapability change

Amazon · 31 AugAWS recognized as a Leader in The Forrester Wave: AI Infrastructure Solutions, Q4 2025Company news