Panga-Azazia/LFM2.5-350M-Base-ASR
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.35BQuant:BF16Context Size:32kPublished:Oct 5, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
Panga-Azazia/LFM2.5-350M-Base-ASR is a 350 million parameter language model developed by Panga-Azazia, finetuned from LiquidAI/LFM2.5-350M-Base. This model was trained using Unsloth and Huggingface's TRL library, achieving a 2x speedup in the training process. It is designed for general language understanding and generation tasks, leveraging its efficient training methodology for practical applications.
Loading preview...
Overview
Panga-Azazia/LFM2.5-350M-Base-ASR is a 350 million parameter language model, developed by Panga-Azazia. It is finetuned from the LiquidAI/LFM2.5-350M-Base model and licensed under Apache 2.0.
Key Capabilities
- Efficient Training: This model was trained 2x faster using Unsloth and Huggingface's TRL library, highlighting an optimized training approach.
- Base Language Model: As a base model, it provides foundational language understanding and generation capabilities, suitable for various downstream tasks.
Good For
- Rapid Prototyping: Its efficient training process makes it suitable for developers looking to quickly experiment with and deploy language models.
- General NLP Tasks: Can be adapted for a range of natural language processing applications, leveraging its base model architecture.
- Resource-Efficient Deployment: With 350 million parameters, it offers a balance between performance and computational resource requirements, making it viable for environments with limited resources.