IBM drops open-weight Granite 4.2 family with built-in agentic capabilities under Apache 2.0
decoder ·
IBM has released its Granite 4.2 language model family in 3B, 8B, and 30B parameter sizes, trained from scratch on about 15 trillion tokens. The models support context windows up to 512,000 tokens and can switch between thinking and non-thinking modes, with a low-effort option for simple queries. The 8B and 30B versions underwent agentic reinforcement learning, teaching them to use tools, write and run code, and search the web in sandbox environments. All models support OpenAI-format tool calling and run on vLLM or SGLang. IBM also introduced Granite Speech 5.0 Turbo CTC models with 470 million parameters, which are twice as fast as previous leaders on the Open ASR Leaderboard and can transcribe three hours of audio per second. All models are released under the Apache 2.0 license on platforms such as Hugging Face, Ollama, and GitHub.
decoder