leokmax/zeta-sft
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 11, 2025License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
leokmax/zeta-sft is a 0.5 billion parameter causal language model developed by leokmax, fine-tuned from unsloth/qwen2.5-coder-7b-bnb-4bit. It was trained using Unsloth and Huggingface's TRL library, enabling 2x faster fine-tuning. This model is optimized for tasks benefiting from efficient training and a compact size, offering a context length of 32768 tokens.
Loading preview...
Overview
leokmax/zeta-sft is a 0.5 billion parameter language model developed by leokmax, fine-tuned from the unsloth/qwen2.5-coder-7b-bnb-4bit base model. This model leverages the Unsloth library and Huggingface's TRL library, which facilitated a 2x faster training process.
Key Capabilities
- Efficient Fine-tuning: Benefits from Unsloth's optimizations for faster training.
- Compact Size: At 0.5 billion parameters, it offers a smaller footprint suitable for resource-constrained environments.
- Extended Context: Supports a context length of 32768 tokens, allowing for processing longer sequences of text.
Good For
- Applications requiring a lightweight yet capable language model.
- Scenarios where rapid fine-tuning on custom datasets is crucial.
- Tasks that can benefit from a model with a substantial context window.