Qwen2 14.77B configuration
Captured config identifies Qwen2ForCausalLM and qwen2, with Safetensors metadata reporting 14770033664 parameters.
Open Source Model Profile · BallAdMyFi
qwen-desi-v1
qwen-desi-v1 is a 14.77B-parameter Qwen2 text-generation fine-tune from BallAdMyFi. Its model card identifies qwen-desi as the source and Unsloth plus TRL as the training stack.
- Publisher
- BallAdMyFi
- Task
- text-generation
- Model type
- qwen2
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
qwen-desi-v1 is published by BallAdMyFi as a qwen2-based text-generation fine-tune. The captured configuration identifies Qwen2ForCausalLM and Safetensors metadata reports 14770033664 parameters, with card data recording apache-2.0. According to the model card, it was finetuned from qwen-desi using Unsloth and Hugging Face TRL.
Recorded capabilities
Documented qwen-desi source
According to the model card, the model was finetuned from qwen-desi and developed by BallAdMyFi.
Unsloth and TRL stack
According to the model card, training used Unsloth with Hugging Face TRL, described as 2x faster.
Apache-2.0 conversational record
Card data records apache-2.0, and hub tags include conversational, text-generation-inference, unsloth, trl, and endpoints-compatible.
Use cases in the source record
- Conversational text generation using the captured transformers Qwen2 configuration with inference-endpoint compatibility.
- Unsloth and TRL fine-tune research building on the documented qwen-desi source model.
Limitations and unknowns
- No evaluation results were extracted from this record.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
- Source-model and training-speed claims come from a brief publisher model card and were not independently verified by Ethen.
Source and provenance
Source: BallAdMyFi/qwen-desi-v1
Captured: Unknown. Processed: 2026-09-07T19:35:34.128598+00:00.
Uploaded model Developed by: BallAdMyFi License: apache-2.0 Finetuned from model : qwen-desi This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.
F001F002F003F004F005F006F007F008F009F010F011