pratinavseth/cricket-captain-qwen3-06b-merged

TEXT GENERATIONConcurrency Cost:1Model Size:0.8BQuant:BF16Ctx Length:32kPublished:Apr 26, 2026License:mitArchitecture:Transformer Open Weights Cold

The pratinavseth/cricket-captain-qwen3-06b-merged model is a 0.8 billion parameter language model based on Qwen/Qwen3-0.6B, with a context length of 32768 tokens. It integrates a LoRA adapter (pratinavseth/cricket-captain-qwen3-06b-stage2) directly into its base weights, making it a single-file model ready for direct use without PEFT. This model is specifically fine-tuned for tasks related to the 'cricket-captain' prompt schema, indicating its specialization in a particular domain or environment.

Loading preview...

Overview

The pratinavseth/cricket-captain-qwen3-06b-merged model is a specialized language model built upon the Qwen/Qwen3-0.6B architecture. It features 0.8 billion parameters and supports a substantial context length of 32768 tokens. A key characteristic of this model is the direct integration of the pratinavseth/cricket-captain-qwen3-06b-stage2 LoRA adapter into its base weights. This merging process results in a single-file model that can be loaded and used directly with standard libraries like transformers, vllm, or TGI, eliminating the need for PEFT (Parameter-Efficient Fine-Tuning) during deployment.

Key Capabilities

  • Direct Deployment: Ready for immediate use without additional PEFT configuration.
  • Specialized Prompt Schema: Designed to operate with the 'cricket-captain' prompt schema, indicating a focus on a specific domain or application.
  • Qwen3-0.6B Base: Leverages the foundational capabilities of the Qwen3-0.6B model.
  • High Context Length: Supports a 32768-token context, beneficial for tasks requiring extensive input understanding.

Good For

  • Applications requiring a compact yet specialized language model.
  • Use cases within the 'cricket-captain' domain or environments that utilize its specific prompt schema.
  • Developers seeking a pre-merged, single-file model for simplified deployment and inference.