CUSTOM
CLANKERS.
← Browse all models

Delta-Vector / Model details

Nanuq-R1-14B

Delta-Vector/Nanuq-R1-14B

fine-tuned
Download files on Hugging Face ↗

Read the original model card ↗ · Choosing a model

Family: Unknown. No supported family evidence recorded.

About this model

Delta-Vector’s Nanuq-R1 14B is a safetensors text-generation model derived from snwy/frankenqwen3-8B-235B-dense-conversion-interleaved-untuned. The creator describes it as a full post-training heal of Snwy’s Frankenmerge between Q3 235B and Q3 8B, intended as a testing ground for GRPO experiments. According to the model card, it received two epochs of pretraining on 1B tokens of creative-writing data, followed by SFT on the creator’s and Pocketdoc’s instruct dataset and GRPO with the Claude-2.7K dataset using Verifiers, with the stated aim of making it more like Claude. The creator says the reward signal tended to flatten after warm-up but judged the model decent and released it as a starting point. The card says it was tuned for ChatML formatting; no license is listed in the metadata.

Description sources

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model
snwy/frankenqwen3-8B-235B-dense-conversion-interleaved-untuned
Parameters
13.5932B
Main repository formats
safetensors
License tag
Unknown
Languages
Unknown
Last updated
2025-09-28 · 03:31:17.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

  • fine-tunedsource tag: base_model:finetune:snwy/frankenqwen3-8B-235B-dense-conversion-interleaved-untuned

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

  • No supported family evidence recorded.

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/delta-vector-nanuq-r1-14b/)

HTML

<a href="https://customclankers.com/models/delta-vector-nanuq-r1-14b/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Tools to customize it further

Remove more refusals

  • Heretic ↗ ★ 36,457

    Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.

  • OBLITERATUS ↗ ★ 8,949

    A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.

Fine-tune it on your own data

  • Unsloth ↗ ★ 78,549

    A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.

  • LlamaFactory ↗ ★ 75,720

    Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.

Make a smaller download (quantize)

  • llama.cpp ↗ ★ 132,313

    Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.

  • MLX LM ↗ ★ 7,470

    Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.

Merge it with another model

  • mergekit ↗ ★ 7,415

    Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.

Learn the techniques

  • LLM Course ↗ ★ 84,684

    A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads
395
Likes
4
Metadata checked
2026-10-26 · 08:12:28 UTC
First seen in index
2026-10-26 · 08:12:28 UTC
Source revision
af2eb93aed72a6120c7ea4e0d160f0318150a4d7

Source-card excerpt

Source model card excerpt: Nanuq-R1 14B Nanuq-R1 14B Created by Delta-Vector → Model Information Nanuq-R1 14B GRPO Experiment Q3-235B-8B Merge/heal Competant assistant with decent writing! A sequel! The

From Delta-Vector/Nanuq-R1-14B. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗