BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

Granite 4.1 30B Instruct

By IBM · United States

chat general code multilingual
Parameters
30B
License
Apache 2.0
Context
128k
VRAM (Q4)
17 GB
Released
29 April 2026

Overview

IBM's dense 30B Granite 4.1: Apache 2.0, 12 languages, 131k context, with OpenAI-compatible tool calling. Built on the same GB200 NVL72 cluster as the rest of the 4.1 lineup.

When to pick this model

  • Enterprise agents requiring OpenAI-compatible function calling
  • Apache 2.0 deployments where Granite 8B isn't enough
  • Multilingual products across 12 languages including French
  • Long-context workflows up to 131k
  • Single-GPU production on RTX 5090 or A100 class hardware

VRAM requirements by quantization

QuantizationVRAM required
Q4_K_M (recommended)17 GB
Q5_K_M21 GB
Q8_032 GB
FP16 (no quantization)60 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Strengths

  • Apache 2.0 with IBM-grade transparency
  • Native OpenAI function-calling schema
  • 12 languages including French
  • 131k context window
  • Official Ollama tag with multiple quantizations

Limitations

  • Needs ~32 GB VRAM at Q4 — RTX 5090 territory
  • No MoE variant at this size
  • Non-English reasoning trails English

Architecture & training

Architecture: Dense Transformer · 64 layers · GQA 32Q/8KV · embedding 4096 · MLP hidden 32,768 · SwiGLU · RoPE · RMSNorm

Training: Fine-tuned from Granite-4.1-30B-Base. SFT + RL alignment pipeline. 12 languages: EN, DE, ES, FR, JA, PT, AR, CS, IT, KO, NL, ZH. NVIDIA GB200 NVL72 cluster (CoreWeave).

Verdict

The Granite to pick when 8B feels light: Apache 2.0, function-calling native, and built for enterprise.

Quick start

ollama run granite4.1:30b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Tools

Is Granite 4.1 30B Instruct the right pick for you?

Compute self-hosted ROI → Back to catalog