Granite 3.3 8b instruct

Model Overview

Granite-3.3-8B-Instruct model is fine-tuned for improved reasoning and instruction-following capabilities.

This model is designed to handle general instruction-following tasks and can be integrated into AI assistants across various domains, including business applications.

• Model Architecture: Granite-3.3-8B-Instruct is a 8-billion parameter 128K context length language model.
• Website: Granite Docs
• Model Source: ibm-granite/granite-3.3-8b-instruct
• Release Date: April 16th, 2025
• License: Apache 2.0
• Supported Languages: English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. However, users may finetune this Granite model for languages beyond these 12 languages.

QPC Configurations

Precision	SoCs / Tensor slicing	NSP-Cores (per SoC)	Full Batch Size	Chunking Prompt Length	Context Length (CL)	QPC URL	QPC Size	QPC Download	Onnx URL	Onnx Download	Generation Date
MXFP6	2	16	1	128	8192	https://dc00tk1pxen80.cloudfront.net/SDK1.21.2/ibm-granite/granite-3.3-8b-instruct/ibm-granite_granite-3.3-8b-instruct_qpc_16cores_128pl_8192cl_1fbs_2devices_mxfp6_mxint8_ccl.tar.gz	7.7GB	Download	https://dc00tk1pxen80.cloudfront.net/SDK1.21.2/ibm-granite/granite-3.3-8b-instruct/ibm-granite_granite-3.3-8b-instruct_ONNX.tar.gz	Download	17-Mar-2026

Run This Model

Download QPCs

mkdir -p ibm-granite/granite-3.3-8B-Instruct
cd ibm-granite/granite-3.3-8B-Instruct

# Download QPC
wget <QPC_Download_URL>
tar xzvf <qpc_filename.tar.gz>

Run QPC

Replace QPC_PATH with actual extracted QPC directories.

python3 -m vllm.entrypoints.openai.api_server \
  --port <PORT> \
  --model ibm-granite/granite-3.3-8B-Instruct \
  --device-group <DEVICE_IDS> \
  --max-model-len <CTX_LEN> \
  --max-seq-len-to-capture <PREFILL_SEQ_LEN> \
  --max-num-seqs <MAX_NUM_SEQS> \
  --quantization mxfp6 \
  --kv-cache-dtype mxint8 \
  --override-qaic-config "num_cores:[num_cores] qpc_path:[qpc_path] ccl_enabled:True comp_ctx_lengths_prefill=[1024,2048,4096,8192] comp_ctx_lengths_decode=[1024,2048,4096,8192]"

Run Inference

Once the server is running, send a request to the OpenAI-compatible endpoint:

curl http://localhost:<PORT>/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ibm-granite/granite-3.3-8B-Instruct",
    "messages": [
      {"role": "user", "content": "<PROMPT>"}
    ],
    "max_tokens": 200
  }'

Useful Links

• Environment Setup