DeepScaleR

DeepScaleR

Agentica Project
Qwen3.8-27B

Qwen3.8-27B

Alibaba
+
+

Related Products

  • LeaseAccounting.app
    Visit Website
  • BYDFi
    5,934 Ratings
    Visit Website
  • Runpod
    230 Ratings
    Visit Website
  • PackageX OCR Scanning
    48 Ratings
    Visit Website
  • Setplex
    10 Ratings
    Visit Website
  • Criminal IP ASM
    20 Ratings
    Visit Website
  • LTX
    182 Ratings
    Visit Website
  • Evertune
    1 Rating
    Visit Website
  • Eurekos
    89 Ratings
    Visit Website
  • Dynamo Software
    71 Ratings
    Visit Website

About

DeepScaleR is a 1.5-billion-parameter language model fine-tuned from DeepSeek-R1-Distilled-Qwen-1.5B using distributed reinforcement learning and a novel iterative context-lengthening strategy that gradually increases its context window from 8K to 24K tokens during training. It was trained on ~40,000 carefully curated mathematical problems drawn from competition-level datasets like AIME (1984–2023), AMC (pre-2023), Omni-MATH, and STILL. DeepScaleR achieves 43.1% accuracy on AIME 2024, a roughly 14.3 percentage point boost over the base model, and surpasses the performance of the proprietary O1-Preview model despite its much smaller size. It also posts strong results on a suite of math benchmarks (e.g., MATH-500, AMC 2023, Minerva Math, OlympiadBench), demonstrating that small, efficient models tuned with RL can match or exceed larger baselines on reasoning tasks.

About

Qwen3.8-27B is an announced 27-billion-parameter model in Alibaba’s Qwen3.8 family, positioned as the compact open-weight counterpart to the much larger Qwen3.8-Max. Qwen introduced the broader Qwen3.8 generation as a new frontier model family focused on coding, agentic work, multimodal understanding, and long-running autonomous tasks. The 27B release is intended to bring that generation to a size that is far more practical for local deployment, experimentation, fine-tuning, and integration into developer workflows. Qwen has confirmed that Qwen3.8-27B will be released with open weights, extending the company’s line of downloadable mid-sized models for users who want direct control over inference and deployment. At the time of announcement, Qwen had not yet published the model card, benchmark table, architecture details, context length, quantization options, or complete deployment guidance for the 27B checkpoint.

Platforms Supported

Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook

Platforms Supported

Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook

Audience

Researchers, students, and developers interested in an AI model capable of mathematical reasoning and logic tasks without requiring heavy hardware

Audience

Local AI application developers who want a compact open-weight model for coding, agents, multimodal work, and self-hosted experimentation

Support

Phone Support
24/7 Live Support
Online

Support

Phone Support
24/7 Live Support
Online

API

Offers API

API

Offers API

Screenshots and Videos

Screenshots and Videos

Pricing

Free
Free Version
Free Trial

Pricing

No information available.
Free Version
Free Trial

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Reviews/Ratings

Overall 0.0 / 5
ease 0.0 / 5
features 0.0 / 5
design 0.0 / 5
support 0.0 / 5

This software hasn't been reviewed yet. Be the first to provide a review:

Review this Software

Training

Documentation
Webinars
Live Online
In Person

Training

Documentation
Webinars
Live Online
In Person

Company Information

Agentica Project
Founded: 2025
United States
agentica-project.com

Company Information

Alibaba
Founded: 1999
China
qwen.ai

Alternatives

DeepCoder

DeepCoder

Agentica Project

Alternatives

Qwen3.6

Qwen3.6

Alibaba
Phi-4-reasoning

Phi-4-reasoning

Microsoft
Qwen2

Qwen2

Alibaba
Qwen3.5

Qwen3.5

Alibaba
Athene-V2

Athene-V2

Nexusflow
Qwen3.8-Max

Qwen3.8-Max

Alibaba

Categories

Categories

Integrations

Alibaba Cloud
Alibaba Cloud Model Studio
Cherry Studio
Cline
ClinePass
Hermes Agent
Hugging Face
Model Context Protocol (MCP)
ModelScope
Novita AI
Odysseus
OfoxAI
Ollama
OpenClaw
Python
Qwen
Qwen Code
Qwen Studio
QwenCloud

Integrations

Alibaba Cloud
Alibaba Cloud Model Studio
Cherry Studio
Cline
ClinePass
Hermes Agent
Hugging Face
Model Context Protocol (MCP)
ModelScope
Novita AI
Odysseus
OfoxAI
Ollama
OpenClaw
Python
Qwen
Qwen Code
Qwen Studio
QwenCloud
Claim DeepScaleR and update features and information
Claim DeepScaleR and update features and information
Claim Qwen3.8-27B and update features and information
Claim Qwen3.8-27B and update features and information