alt.freeweights.granite / granite-4.0-h-small-FP8

From: ibm-granite <huggingface.co/ibm-granite>Newsgroups: alt.freeweights.graniteSubject:

granite-4.0-h-small-FP8

Date: 2026-11-15Commit: 9e3273dSigned: 2DC28CC7B26A283B ✓

Text generation model with 32.6B parameters. Safetensors weights, 33.5 GB in all. Published by ibm-granite under Apache-2.0.

32.6B params33.5GBgranite-4.0-h-small-FP8safetensorspermissive

FactsParams32.6BSize33.5 GBFormatssafetensorsTasktext generationLicenseApache-2.0From sourcepermissive

$ fw get ibm-granite/granite-4.0-h-small-FP8
# manifest sha256 7e84994c3216… · 33.5 GB · 20 files
$ fw verify ./granite-4.0-h-small-FP8 ibm-granite/granite-4.0-h-small-FP8

From sourceHosted (signed)ibm-granite/granite-4.0-h-small-FP8Last check: 2026-11-15 05:15 UTC

Run at home Get help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Quantized of granite-4.0-h-small.

Works with

Run24 Fine-tune29 Convert15 Evaluate13 Scan13 Apps58

Runs with

Ranked by fit to this model (safetensors), then by GitHub stars.

vLLM exo Text Generation WebUI

SGLang MLC LLM KTransformers LMCache Xinference LMDeploy mistral.rs Mooncake FlashInfer

Show all 24

AIBrix fastllm ExLlamaV3

Transformers LocalAI tinygrad gpu-unlock Triton Inference Server BentoML CTranslate2 ExLlamaV2 LitServe

Convert with

mergekit TensorRT Model Optimizer

LLM Compressor Neural Compressor AutoRound AQLM GPTQModel ONNX Core ML Tools ExecuTorch

Show all 15

Evaluate with

Opik DeepEval Ragas lm-evaluation-harness OpenCompass SWE-bench

simple-evals TruLens HELM

LightEval

Evaluate Weave

Show all 13

Autoevals

Scan with

promptfoo garak Guardrails NeMo Guardrails Giskard DeepTeam AI Fairness 360 Fairlearn Detoxify MCP Scanner AgentDojo ModelScan

Show all 13

picklescan

Overview Model card Files Use it License Provenance Model tree

Overview

Size: 33.5 GB
Weights: 33.5 GB (safetensors)
Parameters: 32.6B
Formats: safetensors
Task: text generation
License: Apache-2.0 (permissive)
Files: 20, each with a sha256
Publisher: ibm-granite
Added: 2026-11-15
Section: alt.freeweights.granite

Lineage

Original release from the publisher.

Base model: ibm-granite/granite-4.0-h-small

Where these bytes come from

Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources

Badge	Host	Last check	Result
From source	huggingface.co	2026-11-15 05:15 UTC	passed

Model card

From the publisher at commit 9e3273d. Shown as plain text with links; images open at the publisher.

Granite 4.0 H-Small (FP8)#

📣 Update [10-07-2025]: Added a default system prompt to the chat template to guide the model towards more professional, accurate, and safe responses.

Note. This repository contains the FP8 version of Granite-4.0-H-Small.

Please refer to the the original instruct model's model card for additional details: https://huggingface.co/ibm-granite/granite-4.0-h-small

Files

20 files · 33.5 GBWeights 33.5 GB (safetensors). Sizes and sha256 from the signed manifest.

Hide all 20 files with their sha256

File	Size	sha256
.gitattributes	1.52 KB	11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md	561 B	227abe78c64f28b0ca34754ec50793a0743066f555473cba807002370126f6dc
chat_template.jinja	6.42 KB	9524df67b77a7b25a2dfee898f75b316a157eb9d855b51e32aeac79d7c8a83ce
config.json	5.22 KB	a2cfc41ea2bfa6cf2b4c1e424d78242fbbdad707d361e1672481592300160dd9
generation_config.json	147 B	73f71d577837e2519414d6877b7f5acaeaba3eba6b79a447f46c6a741e9f7b58
merges.txt	917 KB	b6fe424e334903f7fb84d3a106d9730455f4744b9fe3c21ee136d97a00e72502
model-00001-of-00007.safetensors	4.83 GB	bbb7368ef6d609535e9b4ffa6dd16842acdf5da2ed875d9b0a444ce952cfadbb
model-00002-of-00007.safetensors	4.75 GB	3741133a302d8f7b8a9a1a1e6c7cf4944be09db04fae05c16bb8b6e6436a0094
model-00003-of-00007.safetensors	4.75 GB	430a1f453e31beac9a1e5e826f36e00cf9b15a446fac003dfa5e4e4817d576d8
model-00004-of-00007.safetensors	4.81 GB	e7f238cb345de71046cd3e9fefc9b91e8ffb827215496ff050b213feab6d19f0
model-00005-of-00007.safetensors	4.75 GB	36947829bc61e57a681cc42c1cdc10b7f3677d9ce07f70571ebfb8efbb77ae67
model-00006-of-00007.safetensors	4.81 GB	51b7e2e41fd985a6409bd6c6549610b08d45aa93ff8618a4fb6046f3a00c661f
model-00007-of-00007.safetensors	4.77 GB	05105bd1671772a7a4b991f1d391240079a54840ac0be99ad923d82620af7747
model.safetensors.index.json	72.5 KB	7849f1e5bbcff5dd4764b4e3c7079b26973b39566f84ff9f25a9e4195e2f1d3f
model.sig	10.3 KB	ed1d411dffff404d80b0991477f5ebde214fa86f8a376c911de7bf297c4ecd8f
recipe.yaml	207 B	ab883b591568ac1866d1a77148fabf48df8f2d636a38db65021d70895b8a857d
special_tokens_map.json	579 B	c08676c49fd7969a3130f72be6d4bf34da66aa484a6e21dffe359893a1bd5f2e
tokenizer.json	7.15 MB	e2bad66439538cb4d5a7580680932432ed9ece9d3b8577e675512bdf11599253
tokenizer_config.json	17.7 KB	a5ec5daab12ba090a90f3dd169c8f9c275557013a87b9c1258dc7cb497a35c86
vocab.json	1.61 MB	8af71076de8b0b626eed0f4c984faf0a7c062479164b2a31308a948524d4f69c

VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.

Upstream commit	Date	Contents
9e3273d970 pinned now	2026-11-15 05:15 UTC	20 files · 33.5 GB

Use it

$ fw get ibm-granite/granite-4.0-h-small-FP8
# manifest sha256 7e84994c3216… · 33.5 GB · 20 files
$ fw verify ./granite-4.0-h-small-FP8 ibm-granite/granite-4.0-h-small-FP8

fw.pyChecks the signature and every sha256, tries each source in order, resumes.

$ python3 fw.py get https://freeweights.allternit.com/manifests/ibm-granite/granite-4.0-h-small-FP8.json --dir ./granite-4.0-h-small-FP8
$ python3 fw.py verify ./granite-4.0-h-small-FP8 ./ibm-granite__granite-4.0-h-small-FP8.json

hfThe Hugging Face CLI with one setting. Same repo id.

$ HF_ENDPOINT=https://freeweights.allternit.com hf download ibm-granite/granite-4.0-h-small-FP8 --revision 9e3273d970f0dfb1425b755c12378d937762a3c2 --local-dir ./granite-4.0-h-small-FP8

transformersLoad the weights in Python.

import os
os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com"
from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("ibm-granite/granite-4.0-h-small-FP8", revision="9e3273d970f0dfb1425b755c12378d937762a3c2")
model = AutoModelForCausalLM.from_pretrained("ibm-granite/granite-4.0-h-small-FP8", revision="9e3273d970f0dfb1425b755c12378d937762a3c2", device_map="auto")

vLLMServe an OpenAI-compatible API.

$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve ibm-granite/granite-4.0-h-small-FP8 --revision 9e3273d970f0dfb1425b755c12378d937762a3c2

Run at home

License

Name: Apache License 2.0
Id: Apache-2.0
Class: Permissive
Redistribution: yes
Listing: Hosted (signed)
Decided: License as declared by the publisher. No person read it. 2026-11-15
Text in: README.md
Text sha256: 227abe78c64f28b0ca34754ec50793a0743066f555473cba807002370126f6dc
Terms gate: No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: ibm-granite <huggingface.co/ibm-granite>Newsgroups: alt.freeweights.graniteSubject: granite-4.0-h-small-FP8Date: 2026-11-15Commit: 9e3273dSigned: 2DC28CC7B26A283B ✓

Upstream: ibm-granite/granite-4.0-h-small-FP8
Pinned commit: 9e3273d970f0dfb1425b755c12378d937762a3c2
Manifest: granite-4.0-h-small-FP8.json · .minisig
Manifest sha256: 7e84994c3216c55a52f557d869e6911f5ca81f8ad5c1a2203a8efa604f142504
Signature: minisign, Ed25519, key 2DC28CC7B26A283B
Created: 2026-11-15 05:15 UTC
Generator: freeweights-manifest/0.1.0
Takedowns: None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/ibm-granite/granite-4.0-h-small-FP8.json
$ curl -O https://freeweights.allternit.com/manifests/ibm-granite/granite-4.0-h-small-FP8.json.minisig
$ minisign -Vm granite-4.0-h-small-FP8.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq
$ shasum -a 256 <file>   # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

Parent

ibm-granite/granite-4.0-h-small · this model is a quantization of it

No model in the catalog names this one as its parent.

Get it

Download

$ HF_ENDPOINT=https://freeweights.allternit.com hf download ibm-granite/granite-4.0-h-small-FP8

Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest

minisign, key 2DC28CC7B26A283B. granite-4.0-h-small-FP8.json. Provenance. Takedown history: none.

Source

huggingface.co/ibm-granite/granite-4.0-h-small-FP8

Related models

All of alt.freeweights.granite →

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.Edu.seed1

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.Edu.seed2

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.Edu.seed3

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.seed1

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.seed2

From: ibm-granite <huggingface.co/ibm-granite>Subject: GneissWeb.7B_ablation_model_on_350B_FineWeb.seed3

granite-4.0-h-small-FP8

Works with

Runs with

Fine-tune with

Convert with

Evaluate with

Scan with

Works with

Overview

Model card

Granite 4.0 H-Small (FP8)#

Files

Use it

License

Provenance

How to verify

Model tree

Related models