Arete-27B

Released
2026-11-06
Last refreshed
2026-11-09
Status
Researched 1d ago
Open sourceCommercial use: permitted

Arete-27B is released 2026-11-06 in the Arete family with open-source; evaluate it while provider pricing coverage matures.

Nous Research releases · 1 in the last 12 monthsChangelog →
Specifications
Family
Arete
Released
2026-11-06
Parameters
27B
Architecture
Decoder Only
Specialization
general
Openness
Open source
License
Apache 2.0OSI-approvedPermitted
Weights
Available
Code
Unknown
Training
Fine-tuned
Created by
Nous Research

Human-centric AI model innovation

New York, New York, United States
Founded 2023
Website
Pricing

No tracked provider pricing is available yet.

Links
Website HuggingFace

About

Arete-27B is Nous Research's supervised fine-tune of Alibaba's Qwen3.8-27B, released November 6, 2026 with open weights under Apache 2.0. It was post-trained on about 52.8 million text tokens at 64K context, with Inkling as the teacher model, to replace refusals, deflections and state-aligned framing on politically sensitive topics with factual answers.

Providers

No active provider routes are tracked for this model yet.

Capabilities

No model capability flags are currently sourced.

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

NousResearch/Arete-27B