Granite 4.1 8B Instruct

Overview

IBM's dense 8B Granite 4.1 release: Apache 2.0, 12 languages, 131k context, MMLU 73.84, HumanEval 85.37. Trained on a CoreWeave GB200 NVL72 cluster.

When to pick this model

• Enterprise deployments needing Apache 2.0 and IBM provenance
• Tool-calling agents with predictable behavior at moderate scale
• Multilingual products across 12 languages including French
• Long-context tasks up to 131k tokens
• Coding workloads on a single mid-range GPU

VRAM requirements by quantization

Quantization	VRAM required
Q4_K_M (recommended)	5 GB
Q5_K_M	6 GB
Q8_0	9 GB
FP16 (no quantization)	16 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.

Published benchmark scores

Benchmark	Score
MMLU	73.84
GSM8K	92.49
HumanEval	85.37
ArenaHard	68.98
AlpacaEval	50.08

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

Strengths

• Apache 2.0 with full transparency on training
• Strong tool calling and instruction following
• 12 native languages including French
• 131k context window
• Excellent quality-per-parameter at the 8B tier

Limitations

• No official Ollama tag at release
• Reasoning in non-English languages still trails English
• No MoE variant at this size

Architecture & training

Architecture: Dense Transformer · 40 layers · GQA 32Q/8KV · embedding 4096 · MLP hidden 12,800 · RoPE

Training: Improved post-training: SFT + RL alignment. 12 languages: EN, DE, ES, FR, JA, PT, AR, CS, IT, KO, NL, ZH. NVIDIA GB200 NVL72 cluster (CoreWeave).

Quick start

# HuggingFace : ibm-granite/granite-4.1-8b

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