The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestFastino2.5
FastinoText / LLM

Fastino's GLiNER2.5-Decide targets lean NLP tasks

A sub-1B model bundling entity extraction, intent, sentiment and topic classification arrives on Hugging Face.

Sep 23, 2026
UpdateOther
GLiNER2.5-Decide

Fastino has released GLiNER2.5-Decide, a compact text model built for a cluster of everyday natural-language tasks: named-entity recognition, intent detection, sentiment analysis and topic classification. It continues the GLiNER lineage, which has become a popular choice for lightweight, flexible extraction work.

The model falls into the sub-1-billion-parameter bracket, positioning it as an efficient option for teams that need dependable structured outputs without the cost and latency of a large language model. That size makes it a candidate for on-device or high-throughput pipelines where every millisecond and dollar counts.

Why it matters

Much production NLP still comes down to a handful of recurring jobs — pulling entities out of text, routing user requests by intent, or tagging content by topic and tone. A single small model that handles several of these can simplify deployment considerably.

  • Combines NER with intent, sentiment and topic classification
  • Sits under 1B parameters for efficient inference
  • Distributed openly through Hugging Face

The release is listed under a non-standard "other" license, so teams should review the terms on the model page before adopting it. Fastino has not published detailed benchmark figures alongside the listing, leaving hands-on evaluation as the next step for anyone considering it.

Sources

  • fastino/GLiNER2.5-Decide

    Hugging Face

    Visit
OlderBlack Forest Labs brings FLUX to roboticsBlack Forest Labs · Any-to-Any · 2 months agoNewerPrism Brings Joint Video-Audio Generation to DiffusionFrancisRing · Image → Video · 2 months ago

Get the model

Hugging Face

Specs

Size1.9 GB
PrecisionFP32
ArchitectureSpanExtractor
LicenseOTHER
Downloads58.7K
Likes459

Can you run it?

Runs on just about anything.

  • FP32 (as published)

    Any modern laptop · 8 GB RAM

    3.2 GB
  • 16-bit

    Any modern laptop · 8 GB RAM

    2.3 GB
  • 8-bit

    Any modern laptop · 8 GB RAM

    1.8 GB
  • 4-bit

    Any modern laptop · 8 GB RAM

    1.6 GB
Your machine
GGUF builds
  • KV cache approximated from model size (no config metadata) at 8,192 tokens.

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Text / LLM

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More from Fastino

All Fastino releases →
GLiNER 2.5 Multi
Fastino/Text / LLMRuns anywhere

Fastino releases GLiNER 2.5 Multi for extraction

A multilingual model built for structured extraction tasks like named-entity recognition, relation extraction, and JSON output.

Aug 14, 2026

More in Text / LLM

All Text / LLM →
Aion-1.0-Instruct-2.3B
Microsoft/Text / LLMRuns anywhere

Microsoft's Aion 1.0 Instruct is a compact 2.3B chat model

A small, multilingual instruction-tuned model aimed at chat and summarization lands on Hugging Face.

Nov 18, 2026
SenseNova-Loop-9B
SenseTime/Vision-LanguageRuns on a laptop

SenseTime Releases Looped 9B Multimodal Model

The company's new SenseNova-Loop-9B pairs vision-language capabilities with a recurrent looped-transformer design.

Nov 17, 2026
Hunyuan Hy4-flash
Tencent/Vision-LanguageMulti-GPU server

Tencent releases Hunyuan Hy4-flash multimodal MoE

The lightweight, mixture-of-experts model adds vision and text instruction-following to Tencent's open Hunyuan lineup.

Nov 17, 2026