squ11z1/claude-oss-350m
squ11z1/claude-oss-350m is a compact 350 million parameter assistant model designed to emulate a Claude-style conversational tone and interaction pattern. Fine-tuned on approximately 200,000 rows of open-source datasets, it emphasizes assistant behavior, instruction following, and consistent identity. This model is optimized for edge deployment, low-memory experimentation, and lightweight assistant tasks where fast local inference is crucial.
Loading preview...
Overview
Claude OSS 350M is an independent open model project, not an official Anthropic release, focused on delivering a familiar Claude-style feel within a compact, edge-sized model. This 350 million parameter model aims to capture the characteristic Claude-style tone and interaction patterns in a lightweight format suitable for constrained environments.
Key Capabilities
- Claude-style Interaction: Designed to emulate the conversational tone and interaction patterns of Claude models.
- Compact Size: A 350M parameter model, making it suitable for resource-limited settings.
- Fine-tuned for Assistant Behavior: Training focused on assistant behavior, conversational tone, instruction following, and maintaining a consistent identity.
- Open-source Data Training: Fine-tuned on approximately 200,000 rows of open-source datasets from Hugging Face.
Good For
- Edge Deployment: Ideal for running on devices with limited computational resources.
- Low-Memory Experimentation: Suitable for developers and researchers working with memory-constrained setups.
- Lightweight Assistant Tasks: Effective for simple conversational AI and instruction-following applications.
- Fast Local Inference: Enables quick processing directly on local machines.
- Compact Multilingual Interaction: Supports multilingual interactions within its small footprint.