Get the app
← back to catalog · registered 2026-08-22 13:56

criscarleo/starcoder2-15b-instruct-v0.1-abliterated

criscarleo 15B GGUF
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/criscarleo%2Fstarcoder2-15b-instruct-v0.1-abliterated"
Response includes
  • classification m8
  • files 30
  • author_summary 6 models
  • readme_text full
Try next
GET/models?author=criscarleo
All 6 models by criscarleo
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
How we label → About M8 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
82
↑ 3,847% in 90 days
Likes
0
Model age
6mo ago
created 2026-03-26
Downloads over time
Now2.3K→from57↑3,847%
08231.6K2.5K57 on Mar 252.3K on Aug 31MarAprMayJunJulAug
Mar 25 → Aug 31 · 27 snapshots · spans 159 days

Metadata

License
agpl-3.0
Quantizations
Q4_K Q6_K Q8_0
Tags
safetensors gguf starcoder2 license:agpl-3.0 endpoints_compatible region:us conversational

Related

Other by criscarleo
DeepSeek-Coder-V2-Lite-Instruct-abliterated 2K Qwen2.5-Coder-7B-Instruct-abliterated 412 Qwen2.5-Coder-14B-Instruct-abliterated 263 qwen2.5-coder-3b-abliterated 38 qwen2.5-coder-3b-abliterated-basic 23
Total size
96.6 GB
Files
30
Quantizations
4
Registered
2026-08-22 13:56
Last updated on HF
2026-03-31 18:33
Source huggingface.co/criscarleo/starcoder2-15b-instruct-v0.1-abliterated ↗

Files by quantization

Q8_0 1 file 15.8 GB
starcoder2-15b-instruct-v0.1-abliterated-Q8_0.gguf 15.8 GB adbaea27 download
Q6_K 1 file 12.2 GB
starcoder2-15b-instruct-v0.1-abliterated-Q6_K.gguf 12.2 GB 60889e72 download
Q4_K 1 file 9.18 GB
starcoder2-15b-instruct-v0.1-abliterated-Q4_K_M.gguf 9.18 GB a1d65e63 download
Auxiliary files 27 files 59.5 GB
starcoder2-15b-instruct-v0.1-abliterated.gguf 29.7 GB dd57bf3b download
model-00006-of-00017.safetensors 1.79 GB eadfa70d download
model-00008-of-00017.safetensors 1.79 GB bc0b77fb download
model-00010-of-00017.safetensors 1.79 GB 6711df0b download
model-00012-of-00017.safetensors 1.79 GB eb9f758a download
model-00014-of-00017.safetensors 1.79 GB ae655160 download
model-00016-of-00017.safetensors 1.79 GB 7fb6046c download
model-00002-of-00017.safetensors 1.79 GB a90175f5 download
model-00004-of-00017.safetensors 1.79 GB 81c21602 download
model-00007-of-00017.safetensors 1.78 GB aba085a8 download
model-00009-of-00017.safetensors 1.78 GB f7e37035 download
model-00011-of-00017.safetensors 1.78 GB 247e44f7 download
model-00013-of-00017.safetensors 1.78 GB 76d5311d download
model-00015-of-00017.safetensors 1.78 GB 745140da download
model-00005-of-00017.safetensors 1.78 GB 750380b2 download
model-00003-of-00017.safetensors 1.78 GB 7c455e4c download
model-00001-of-00017.safetensors 1.71 GB 7d7ba56b download
model-00017-of-00017.safetensors 1.20 GB 049e9b43 download
tokenizer.json 3.32 MB 6b00cf49 download
model.safetensors.index.json 50.9 KB 2efade53 download
.gitattributes 1.82 KB 6afd566d download
abliteration_metadata.json 1.76 KB e34d1846 download
tokenizer_config.json 1.38 KB b9a06bfc download
config.json 818 B 9949d546 download
README.md 783 B 7265800b download
chat_template.jinja 574 B cf07c7d4 download
generation_config.json 110 B 334f05cd download

README current version from Hugging Face

license: agpl-3.0

Work in Progress - Experimental version

Method & Process

This model was processed using the Obliteratus methodology.
The fine-tuning/transformation was executed via the official notebook provided by [pliny-the-prompter](https://huggingface.co/spaces/pliny-the-prompter/obliteratus

Quantization & Inference Engine

  • Framework: llama.cpp
  • Format: GGUF (v3)
  • Original Authors: Georgi Gerganov and the GGML contributors.

References & Credits

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-03-30Create README.md5866ed1783 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration