alt.freeweights.text / sarashina2.2-0.5b-instruct-v0.1
From: sbintuitions <huggingface.co/sbintuitions>Newsgroups: alt.freeweights.textSubject:

sarashina2.2-0.5b-instruct-v0.1

Date: 2026-11-15Commit: e4b9aacSigned: 2DC28CC7B26A283B ✓

Text generation model with 793M parameters. Safetensors weights, 1.59 GB in all. Published by sbintuitions under MIT.

793M params1.59GBsarashina2.2-0.5b-instruct-v0.1safetensorspermissive
FactsParams793MSize1.59 GBFormatssafetensorsTasktext generationLicenseMITFrom sourcepermissive
$ fw get sbintuitions/sarashina2.2-0.5b-instruct-v0.1
# manifest sha256 9592598c6bf8… · 1.59 GB · 10 files
$ fw verify ./sarashina2.2-0.5b-instruct-v0.1 sbintuitions/sarashina2.2-0.5b-instruct-v0.1
From sourceHosted (signed)sbintuitions/sarashina2.2-0.5b-instruct-v0.1Last check: 2026-11-15 03:23 UTC

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

Fine-tune of sarashina2.2-0.5b.

Works with

Overview

Size
1.59 GB
Weights
1.59 GB (safetensors)
Parameters
793M
Formats
safetensors
Task
text generation
License
MIT (permissive)
Files
10, each with a sha256
Publisher
sbintuitions
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: sbintuitions/sarashina2.2-0.5b

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
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 03:23 UTCpassed

Model card

From the publisher at commit e4b9aac. Shown as plain text with links; images open at the publisher.

sbintuitions/sarashina2.2-0.5b-instruct-v0.1#

Model Summary#

This repository provides Japanese language models trained by SB Intuitions.

Model Details#

  • Model type: Autoregressive Language Model
  • Language(s): Japanese

Evaluation in Japanese and English Tasks#

ModelElyza-tasks-100Japanese MT BenchEnglish MT Bench
Qwen/Qwen2.5-0.5B-instruct1.532.954.98
sarashina2.2-0.5B-instruct-v0.12.384.555.09
Rakuten/RakutenAI-2.0-mini-instruct2.414.495.13
SakanaAI/TinySwallow-1.5B-Instruct2.815.246.31
Qwen/Qwen2.5-1.5B-instruct2.284.066.99
llm-jp/llm-jp-3-1.8b-instruct32.534.624.83
sarashina2.2-1B-instruct-v0.12.885.096.46
google/gemma-2-2b-jpn-it3.025.197.56
Qwen/Qwen2.5-3B-instruct2.995.687.88
llm-jp/llm-jp-3-3.7b-instruct32.794.985.44
sarashina2.2-3B-instruct-v0.13.756.517.71

How to Use#

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, set_seed

# モデルのロード
model_name = "sbintuitions/sarashina2.2-0.5b-instruct-v0.1"
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_name)
chat_pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer)
set_seed(123)

# ユーザーの入力
user_input = [{"role": "user", "content": "こんにちは。あなたの名前を教えて"}]

# モデルによる応答生成
responses = chat_pipeline(
    user_input,
    max_length=50,
    do_sample=True,
    num_return_sequences=3,
)

# 応答を表示
for i, response in enumerate(responses, 1):
    print(f"Response {i}: {response['generated_text']}")

# Response 1: [{'role': 'user', 'content': 'こんにちは。あなたの名前を教えて'}, {'role': 'assistant', 'content': 'Sarashina2と言います。本日のご要件を教えて下さい。'}]
# Response 2: [{'role': 'user', 'content': 'こんにちは。あなたの名前を教えて'}, {'role': 'assistant', 'content': 'こんにちは!私の名前はSarashina2です。今日はどうしましたか?'}]
# Response 3: [{'role': 'user', 'content': 'こんにちは。あなたの名前を教えて'}, {'role': 'assistant', 'content': 'Sarashina2と言います。本日のご要件を教えて下さい。'}]

Limitations#

This model has limited safety training. Therefore, it might generate some meaningless sequences, some inaccurate instances, or biased/objectionable outputs. Before using it, we would like developers to tune models based on human preferences and safety considerations.

License#

MIT License

Files

10 files · 1.59 GBWeights 1.59 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 10 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
LICENSE1.07 KB284353c80e0d52c06e97af62fd40e4dd1d253fa28f5d1e756121ea1d5d441509
README.md4.59 KB658487699f4576a2bca1ace84bba34d79344b291b153a0cab64dfc4a0134d19e
config.json676 B09d5e9f815960ebfd59aa88df9de583ea1899c50662905b1b83ef68dcb9639c9
generation_config.json134 B6ae72a2cf3c6417e6e34ab80537dae142e91a3171d2b27bc6fbc28e4a2460883
model-00001-of-00001.safetensors1.59 GB3d886d5ba826134c408fee95666240ae2dc1372e944b7521c8448d4def08502f
model.safetensors.index.json17.1 KBaeae7de712c66876337f817fee2fe8a71c171b22595a91464417dd75d63e9bc4
special_tokens_map.json968 B30bf8256f9a1eb3287af2a9b7940465e29a38aad4a459a3e773bb3a14bd34c0f
tokenizer.model1.83 MB008293028e1a9d9a1038d9b63d989a2319797dfeaa03f171093a57b33a3a8277
tokenizer_config.json6.42 KBff413f71a6cb94015411448e948f908c70853222a070c04cd63cec2b2292dce2
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
e4b9aacc3f pinned now2026-11-15 03:23 UTC10 files · 1.59 GB

Use it

$ fw get sbintuitions/sarashina2.2-0.5b-instruct-v0.1
# manifest sha256 9592598c6bf8… · 1.59 GB · 10 files
$ fw verify ./sarashina2.2-0.5b-instruct-v0.1 sbintuitions/sarashina2.2-0.5b-instruct-v0.1
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/sbintuitions/sarashina2.2-0.5b-instruct-v0.1.json --dir ./sarashina2.2-0.5b-instruct-v0.1 $ python3 fw.py verify ./sarashina2.2-0.5b-instruct-v0.1 ./sbintuitions__sarashina2.2-0.5b-instruct-v0.1.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download sbintuitions/sarashina2.2-0.5b-instruct-v0.1 --revision e4b9aacc3f644893d0179847946ef6c58d868f29 --local-dir ./sarashina2.2-0.5b-instruct-v0.1
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("sbintuitions/sarashina2.2-0.5b-instruct-v0.1", revision="e4b9aacc3f644893d0179847946ef6c58d868f29") model = AutoModelForCausalLM.from_pretrained("sbintuitions/sarashina2.2-0.5b-instruct-v0.1", revision="e4b9aacc3f644893d0179847946ef6c58d868f29", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve sbintuitions/sarashina2.2-0.5b-instruct-v0.1 --revision e4b9aacc3f644893d0179847946ef6c58d868f29

License

Name
MIT License
Id
MIT
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
LICENSE
Text sha256
284353c80e0d52c06e97af62fd40e4dd1d253fa28f5d1e756121ea1d5d441509
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: sbintuitions <huggingface.co/sbintuitions>Newsgroups: alt.freeweights.textSubject: sarashina2.2-0.5b-instruct-v0.1Date: 2026-11-15Commit: e4b9aacSigned: 2DC28CC7B26A283B ✓
Upstream
sbintuitions/sarashina2.2-0.5b-instruct-v0.1
Pinned commit
e4b9aacc3f644893d0179847946ef6c58d868f29
Manifest
sarashina2.2-0.5b-instruct-v0.1.json · .minisig
Manifest sha256
9592598c6bf8b2d51ad18ab8d32442dd655c47935f4551d8d4f88edca843d7b5
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:23 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/sbintuitions/sarashina2.2-0.5b-instruct-v0.1.json $ curl -O https://freeweights.allternit.com/manifests/sbintuitions/sarashina2.2-0.5b-instruct-v0.1.json.minisig $ minisign -Vm sarashina2.2-0.5b-instruct-v0.1.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

sbintuitions/sarashina2.2-0.5b · this model is a finetune of it

1 model in the catalog name this one as their parent.

Quantizations
1
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download sbintuitions/sarashina2.2-0.5b-instruct-v0.1
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. sarashina2.2-0.5b-instruct-v0.1.json. Provenance. Takedown history: none.

Variants

1 quantized or adapter build of sarashina2.2-0.5b-instruct-v0.1. Each has its own license check and page.

From: laion <huggingface.co/laion>Subject: a2-rl-e2egit_large-50-32B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B