Mergekit SLERP merge
According to the model card, this is a merge of pre-trained language models created with mergekit, using slerp with dtype bfloat16 and base model Hasnonname/Qwen2.5-14B-Wheatear-v0.
Open Source Model Profile · Hasnonname
Qwen2.5-14B-Kebab-v0
Qwen2.5-14B-Kebab-v0 is a 14.77B-parameter Qwen2 text-generation merge from Hasnonname. Its model card describes a creative-writing merge of the Kestrel and Wheatear builds using mergekit.
- Publisher
- Hasnonname
- Task
- text-generation
- Model type
- qwen2
- License
- Unknown
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Qwen2.5-14B-Kebab-v0 is published by Hasnonname as a Qwen2 text-generation model. The captured configuration identifies Qwen2ForCausalLM and Safetensors metadata reports 14,770,033,664 parameters. According to the model card, it merges Hasnonname/Qwen2.5-14B-Wheatear-v0 and Hasnonname/Qwen2.5-14B-Kestrel-v0, with Kestrel described for roleplay and multiturn conversation and Wheatear for creative writing.
Recorded capabilities
Kestrel plus Wheatear combination
According to the model card, the merge includes Hasnonname/Qwen2.5-14B-Wheatear-v0 and Hasnonname/Qwen2.5-14B-Kestrel-v0, combining creative writing with roleplay and multiturn conversation data.
Qwen2 Transformers record
Captured configuration records Qwen2ForCausalLM with model type qwen2 and Transformers support.
Use cases in the source record
- Creative-writing and multiturn roleplay text generation using the merged Kestrel and Wheatear behavior described in the model card.
Limitations and unknowns
- No license value was extracted from this record.
- 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 and provenance
Source: Hasnonname/Qwen2.5-14B-Kebab-v0
Captured: Unknown. Processed: 2026-09-07T19:35:06.098500+00:00.
Qwen2.5-14B-Kebab-v0 merge methodology taken from Aletheia v1 hyperparams slightly modified from Sugarquill v1 dataset consists of creative writing, multiturn RP, and some general assistant tasks. kestrel was RP and multiturn conversations, wheatear was creative writing thanks to the folks in the arli discord (owen and fizz in particular) for helping me with my axolotl config quants available here: Qwen2.5-14B-Kebab-v0-GGUF Writing Samples {{[INPUT]}} Write me a short story without a happy ending. {{[OUTPUT]}} The town of Mornhold had been founded nearly five decades earlier in the midst of an unforgiving stretch of land that most o…
F001F002F003F004F005F006F007F008F009F010F014F015