ermiaazarkhalili/FastContext-4B-SFT_base-SFT-Fable5

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 7, 2026Architecture:Transformer Featherless Exclusive Cold

FastContext-4B-SFT_base-SFT-Fable5 is a 4.0 billion parameter causal language model based on the Qwen3ForCausalLM architecture, developed by ermiaazarkhalili. It is a LoRA fine-tune of Microsoft's FastContext-1.0-4B-SFT, specifically supervised fine-tuned on the private ermiaazarkhalili/Fable-5-Complete-2M-Clean dataset. This model is designed for instruction-following tasks, leveraging its fine-tuning on a dedicated instruction dataset.

Loading preview...

Overview

This model, FastContext-4B-SFT_base-SFT-Fable5, is a 4.0 billion parameter causal language model built on the Qwen3ForCausalLM architecture. It was developed by ermiaazarkhalili through a LoRA (Low-Rank Adaptation) fine-tuning process. The base model for this fine-tune was microsoft/FastContext-1.0-4B-SFT, and it was supervised fine-tuned using the private ermiaazarkhalili/Fable-5-Complete-2M-Clean dataset.

Key Training Details

  • Base Model: microsoft/FastContext-1.0-4B-SFT
  • Architecture: Qwen3ForCausalLM
  • Parameters: 4.0 billion
  • Training Data: ermiaazarkhalili/Fable-5-Complete-2M-Clean (private)
  • Method: LoRA supervised fine-tuning using Unsloth and TRL.
  • Max Sequence Length: 4096 tokens
  • Precision: 4-bit (QLoRA)

Limitations

  • No benchmark evaluation has been conducted; only training-loss observations are available.
  • Inherits biases, knowledge cutoff, and potential failure modes from its base model.
  • Fine-tuned exclusively on a single instruction-following dataset, meaning its behavior outside this distribution is untested.
  • LoRA adapters were merged into the base weights, preventing detachment of the fine-tune.