Granite 4.1 30B Architecture
Back to Granite 4.1 30B in the gallery
IBM
IBM's dense 30B Granite 4.1 model keeps a straightforward long-context GQA transformer stack and improves tool calling, instruction following, and chat behavior through post-training.
Architecture and specifications
- Scale
- 30B parameters
- Context (tokens)
- 131,072
- Vocabulary
- 100,352
- License
- Apache License 2.0
- Date
- 2026-04-29
- Decoder type
- Dense
- Attention
- GQA with RoPE
- Layer mix
- 64 GQA
- KV cache / token (bf16) info
- 256 KiB · High
- Key detail
- Uses a 4096-wide dense stack, 32 query heads, 8 KV heads, SwiGLU MLPs, shared input/output embeddings, and Granite-specific MuP scaling factors.
Training and benchmarks
- Pretraining data
- Approximately 15T tokens Multi-stage pretraining, including long-context extension. Granite 4.1 training details
- Post-training methods
- SFT, RLHF, RL SFT on about 4.1M examples, then multi-domain, chat RLHF, identity, and math RL using GRPO with DAPO loss. Granite 4.1 training details