Ai2's open-weight 8B model that writes cited scientific reports from a research question and retrieved literature
Strands Decider 2B
Open-source 2B decision model from AWS's Strands Agents project that picks options and returns scores with a reliability figure
About Strands Decider 2B
Routing and checking steps inside AI agent workflows are the job of Strands Decider 2B, a small open-source model released on 1 October 2026 by the Strands Agents project, the open-source agent framework from AWS.
A model that chooses instead of writing
A decision model does not generate text. It is given a set of options in the prompt, such as yes or no, one of several languages or a category, and it picks one, or it returns a numeric score. Each answer comes with a reliability figure. The authors of the release say this makes it fast and dependable for routing and validation inside agent systems, and they state plainly that it is not meant for coding, chat or summarization.
How it is built
Decider 2B starts from the Qwen 3.5 2B base model. Its text-generation head is removed and replaced by a small pointer head that scores each option, while the base model is fine-tuned with a LoRA adapter. The released version is the nineteenth iteration of the design.
Speed and openness
According to the project, a local CPU or GPU is enough to run it, and each answer takes only tens of milliseconds, which suits checks that sit in the middle of a larger pipeline. Hugging Face hosts the weights, licensed under Apache 2.0, while GitHub holds the code, the training data and the scripts needed to reproduce it.
What the authors claim
They describe accuracy and calibration as competitive with other decision models, based on their own measurements. A group of Clef decision models appeared on the same day, so teams comparing options in this niche have more than one to test.
TrustList summary
Strands Decider 2B is a compact open-source decision model that the Strands Agents project, part of AWS, released on 1 October 2026. Instead of generating text, it picks among options given in the prompt, such as a yes-or-no call or a choice of category, and it can also return a numeric score with a reliability figure attached. The project's authors say it suits routing and checks inside agent workflows, and is not meant for coding, chat or summarising.
What Strands Decider 2B does
- Option selection: picks one answer from the options given in the prompt.
- Scoring: returns a numeric score with a reliability figure for each answer.
- Local running: the project says it runs on a local CPU or GPU, with answers in tens of milliseconds.
- Open code: weights are on Hugging Face under Apache 2.0, and the code, training data and scripts are on GitHub so results can be reproduced.
Who it works with
Developers building agent workflows that need fast routing and checks.
Key facts
- Base model: Qwen 3.5 2B, with its text-generation head removed and a pointer head added
- Licence: Apache 2.0
- Authors: Marc Brooker, Mike Chambers and Fabio Nonato de Paula
The accuracy and calibration figures are the authors' own measurements.
Summary written by TrustList from the business's own published information, Nov 2026.
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| No benchmark scores yet. Be the first to add one. | ||
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
Strands Decider 2B launches
Request a demo or quote from Strands Decider 2B
Reviews
Write the first review of Strands Decider 2B
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about Strands Decider 2B.
Strands Decider 2B alternatives
Similar listings buyers compare against this one.
Microsoft AI's real-time speech-to-text model for voice agents, transcribing 60 languages with continuous language detection
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Microsoft AI's text-to-speech model for keeping one voice across 23 languages, with voice cloning from short audio
Liquid AI's open-weight decision model, giving yes or no, choice or score answers from text and images
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
Anthropic's smallest and fastest Claude 5.5 model, available on its own API and three cloud platforms
Black Forest Labs' image generation and editing model with bounding-box layouts, multiple reference images and 4K output
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.