Last updated 16 Nov 2026. Sources last checked 16 Nov 2026.
Strands Decider E2B
Strands Agents · Released 15 Oct 2026
Learn how to get itDiscuss on X
A Gemma 4-based open decision model in the Strands Decider family (E2B size). Accepts the same noul, choice, and score question types as Strands Decider 2B, so sizes can be swapped without rewriting questions. Aimed at routing, guardrails, evaluation, and other agent decisions.
- Maker
- Strands Agents
- Release date
- 15 Oct 2026
- First seen
- 15 Oct 2026
- Access
- Adapters and readout heads on Hugging Face under the StrandsAgents organization; run with the strands-decider CLI version 0.2.0 or later. Training and inference code is in the Strands Decider repository. Adapters, readout heads, and code are Apache 2.0; Gemma 4 base weights are downloaded separately.
- Pricing (as publicly stated)
- Open source / self-hosted. No hosted list price was stated in the announcement.
- Modalities
- text
- Question types
- noul, choice, score
- Latency (as stated)
- Strands reported a 102 ms median latency on one NVIDIA H100 80 GB in this release run.
- Official announcement
- https://strandsagents.com/blog/strands-decider-gemma-4-models/
Benchmark claims
These are claims as published by the maker or in the linked official post. This catalog has not independently rerun every benchmark.
On JevBench v1's 231 public tasks, Strands reported 180/231 accuracy (77.9%) and a Brier score of 0.313 for E2B. The team noted that JevBench now uses newer releases, so these numbers should be compared only with runs on the same task set.
What enterprises should know
Cost and pricing
Open source / self-hosted. No hosted list price was stated in the announcement. Figures are what the maker stated in the linked source and may have changed.
Lock-in
The official announcement describes distribution as follows: Adapters and readout heads on Hugging Face under the StrandsAgents organization; run with the strands-decider CLI version 0.2.0 or later. Training and inference code is in the Strands Decider repository. Adapters, readout heads, and code are Apache 2.0; Gemma 4 base weights are downloaded separately. That is the maker's description of how the product is obtained, not an assessment of switching costs.
Security and privacy
The official announcement states that weights were released under the Apache 2.0 license. The announcement we sourced did not state a specific security certification or independent privacy audit for this product.
Maturity and availability
The maker's official announcement is dated 2026-10-15. Access was described as: Adapters and readout heads on Hugging Face under the StrandsAgents organization; run with the strands-decider CLI version 0.2.0 or later. Training and inference code is in the Strands Decider repository. Adapters, readout heads, and code are Apache 2.0; Gemma 4 base weights are downloaded separately. Strands reported a 102 ms median latency on one NVIDIA H100 80 GB in this release run.
Compliance notes
The official announcement we sourced did not state a specific regulatory approval or compliance certification for this product.
What consumers should know
Cost and pricing
The announcement describes a way to obtain or run the model (including, where stated, open weights or self-hosting). Hosted prices, if any, are as follows: Open source / self-hosted. No hosted list price was stated in the announcement.
Lock-in
The announcement describes access as: Adapters and readout heads on Hugging Face under the StrandsAgents organization; run with the strands-decider CLI version 0.2.0 or later. Training and inference code is in the Strands Decider repository. Adapters, readout heads, and code are Apache 2.0; Gemma 4 base weights are downloaded separately. Self-hosting, where it is offered, means you are not limited to the maker's hosted endpoint, but you still depend on the published weights, license, and your own hardware.
Security and privacy
The announcement describes local or self-hosted use. What happens to your data then depends on where you run the software. The announcement we sourced did not state a product-specific consumer privacy audit.
Maturity and availability
The official announcement is dated 2026-10-15. Availability was described as: Adapters and readout heads on Hugging Face under the StrandsAgents organization; run with the strands-decider CLI version 0.2.0 or later. Training and inference code is in the Strands Decider repository. Adapters, readout heads, and code are Apache 2.0; Gemma 4 base weights are downloaded separately. Strands reported a 102 ms median latency on one NVIDIA H100 80 GB in this release run. This summary does not say whether the product is suitable for a particular consumer task.
Compliance notes
The official announcement we sourced did not state consumer-protection certifications specific to this product.
Figures are as stated in the linked source on the date shown. Sources last checked 16 Nov 2026. Names and marks belong to their owners. This page is an independent summary, not a product page from Strands Agents.