bernquant/OpenThinkerAgent-32B-SFT-316

TEXT GENERATIONPricing:Input $0.408 / Cached $0.0816 / Output $1.972Concurrent Unit Cost:2Model Size:32BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 7, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

OpenThinkerAgent-32B-SFT-316 is a 32 billion parameter language model developed by OpenThoughts-Agent, post-trained from Qwen3-32B. It is fine-tuned using a 316-example dataset of agentic task-trajectory pairs, specifically designed for agentic capabilities. This model excels in agent-based tasks, demonstrating improved performance on benchmarks like OpenThoughts-TBLite and Terminal-Bench 2.0 compared to its base model.

Loading preview...

OpenThinkerAgent-32B-SFT-316 Overview

OpenThinkerAgent-32B-SFT-316 is a 32 billion parameter model developed by OpenThoughts-Agent, derived from the Qwen3-32B architecture. This model is specifically fine-tuned for agentic tasks through Supervised Fine-Tuning (SFT) on the unique OpenThoughts-Agent-SFT-316 dataset. This dataset comprises 316 high-quality (task, agent-trajectory) pairs, sourced from platforms like SWE-Smith, StackExchange-SuperUser, StackExchange-Tezos, and IssueTasks, with trajectories generated by GLM-4.7-AWQ and filtered for traces with at least 5 model turns.

Key Capabilities & Performance

This model demonstrates enhanced performance in agentic benchmarks:

  • Improved Agentic Task Execution: Outperforms its base model, Qwen3-32B, on agent-specific benchmarks.
  • Benchmark Scores: Achieves 24.2 on OpenThoughts-TBLite and 13.1 on Terminal-Bench 2.0 (pass@1, mean over 3 stochastic re-runs), significantly higher than the base model's 13.7 and 7.5 respectively.
  • Specialized Training: Benefits from full-parameter SFT using a carefully curated dataset focused on agent trajectories.

When to Use This Model

OpenThinkerAgent-32B-SFT-316 is particularly well-suited for:

  • Developing AI Agents: Ideal for applications requiring models to perform multi-step, agent-like reasoning and task execution.
  • Agentic Workflow Automation: Use cases involving automated problem-solving, code generation, or interactive system control where agentic capabilities are crucial.
  • Research in Agentic AI: Provides a strong foundation for further research and development in agent-based language models.

Finetunes
1 models