For RP & story gen,
no one expected Qwen2.5-14B to be that elegant. Fine-tunings fought to be the cherry on top, till Qwen3 melts the cake itself.
The Oscar winner was contented before being thrown onto the real battlefield.
Words say her wounds are dramatic,
overall too charming to be breathing.

I suppose players got lost in comparisons,
letting Sao10K/14B-Qwen2.5-Kunou-v1 the innocent villain hide behind sometimesanotion/Qwenvergence-14B-v13-Prose-DS the cynosure of all eyes,
not to mention a drifting deepcogito/cogito-v1-preview-qwen-14B actually kills via misfire.

Combination of the three could survive a heroic epic life in fantasy worlds.
In reality she is caught by a...
jumping bullet on the cheek.
"If you want to go deep, don't take the blue pill."
models:
  - model: Sao10K/14B-Qwen2.5-Kunou-v1
  - model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
    parameters:
      density: [0.16, 0.26, 0.36, 0.46, 0.56, 0.46, 0.36, 0.26, 0.16]
      weight: [0.166, 0.496, 0.496, 0.166, 0.166, 0.496, 0.496, 0.166]
  - model: deepcogito/cogito-v1-preview-qwen-14B
    parameters:
      density: [0.56, 0.46, 0.36, 0.26, 0.16, 0.26, 0.36, 0.46, 0.56]
      weight: [0.496, 0.166, 0.166, 0.496, 0.496, 0.166, 0.166, 0.496]
merge_method: breadcrumbs
base_model: Sao10K/14B-Qwen2.5-Kunou-v1
parameters:
    gamma: 0.06
    lambda: 0.96
tokenizer_source: base
dtype: bfloat16
Downloads last month
12
Safetensors
Model size
14.8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for AIgotahole/Queen-2.5-14B-aka

Base model

Qwen/Qwen2.5-14B
Finetuned
(1)
this model
Quantizations
3 models

Collection including AIgotahole/Queen-2.5-14B-aka