README.md

metadata

license: apache-2.0
language:
  - en
  - zh
base_model:
  - unsloth/JanusCoder-8B
  - TeichAI/Nemotron-Cascade-8B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill
  - nightmedia/Qwen3-8B-Element
pipeline_tag: text-generation
library_name: transformers
tags:
  - coding
  - research
  - deep thinking
  - 128k context
  - Qwen3
  - All use cases
  - creative
  - creative writing
  - fiction writing
  - plot generation
  - sub-plot generation
  - story generation
  - scene continue
  - storytelling
  - fiction story
  - science fiction
  - all genres
  - story
  - writing
  - vivid prosing
  - vivid writing
  - fiction
  - roleplaying
  - bfloat16
  - finetune
  - mergekit
  - merge
  - mlx

JanusCoder-8B-Nemotron-Claude-Opus-qx86-hi-mlx

Qwen3-8B-Element-qx86-hi-mlx

This model is a 1.4/0.6 nuslerp merge of:

• unsloth/JanusCoder-8B
• TeichAI/Nemotron-Cascade-8B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill

The Nemotron-Cascade base is prone to looping, mainly for the lack of social skills: the addition of just Claude thinking traces without a body of evidence made the Element very smart, but unstable, even with Janus help.

Brainwaves

          arc   arc/e boolq hswag obkqa piqa  wino
qx86-hi   0.532,0.746,0.846,0.738,0.456,0.794,0.709

Janus     0.537,0.731,0.862,0.697,0.446,0.782,0.667
Element   0.532,0.746,0.846,0.738,0.456,0.794,0.709

Perplexity
qx86-hi   4.744 ± 0.036
qx64-hi   4.798 ± 0.036
mxfp4     5.012 ± 0.038

-G

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("JanusCoder-8B-Nemotron-Claude-Opus-qx86-hi-mlx")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True, return_dict=False,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)