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

noneUsername/Seed-Coder-8B-Instruct-abliterated-W8A8

noneUsername Llama 7.0B second-order
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/noneUsername%2FSeed-Coder-8B-Instruct-abliterated-W8A8"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 25
  • author_summary 17 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
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=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
25
13 last 30d - active
Likes
0
Model age
17mo ago
created 2025-05-17
Downloads over time
Now32→from13↑146%
012233513 on May 14, 202532 on Oct 12May '25Aug '25Nov '25FebMayAug
May 14, 2025 → Oct 12 · 114 snapshots · spans 516 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

Tags
safetensors llama 8-bit compressed-tensors region:us

Related

Total size
8.87 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-05-17 18:28

Files by quantization

Auxiliary files 11 files 8.88 GB
model-00001-of-00002.safetensors 4.65 GB 3813eb8d download
model-00002-of-00002.safetensors 4.22 GB 783ec240 download
tokenizer.json 11.3 MB e78ef3f1 download
model.safetensors.index.json 42.4 KB f41e078b download
tokenizer_config.json 27.6 KB ddcb6095 download
README.md 9.90 KB 617325d1 download
config.json 1.75 KB e0bad2bb download
.gitattributes 1.53 KB 52373fe2 download
special_tokens_map.json 613 B cf749aa4 download
recipe.yaml 228 B d0d86359 download
generation_config.json 111 B 59f507fa download

README current version from Hugging Face


base_model:

  • huihui-ai/Seed-Coder-8B-Instruct-abliterated

vllm (pretrained=/root/autodl-tmp/Seed-Coder-8B-Instruct-abliterated,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.552 ± 0.0315
strict-match 5 exact_match ↑ 0.552 ± 0.0315

vllm (pretrained=/root/autodl-tmp/Seed-Coder-8B-Instruct-abliterated,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.566 ± 0.0222
strict-match 5 exact_match ↑ 0.564 ± 0.0222

vllm (pretrained=/root/autodl-tmp/Seed-Coder-8B-Instruct-abliterated,add_bos_token=true,max_model_len=3048,dtype=bfloat16,model_impl=transformers,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups Version Filter n-shot Metric Value Stderr
mmlu 2 none acc ↑ 0.4316 ± 0.0167
- humanities 2 none acc ↑ 0.4205 ± 0.0344
- other 2 none acc ↑ 0.4615 ± 0.0356
- social sciences 2 none acc ↑ 0.4278 ± 0.0359
- stem 2 none acc ↑ 0.4211 ± 0.0289

vllm (pretrained=/root/autodl-tmp/80-128-4096,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.544 ± 0.0316
strict-match 5 exact_match ↑ 0.540 ± 0.0316

vllm (pretrained=/root/autodl-tmp/80-256-4096,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.56 ± 0.0315
strict-match 5 exact_match ↑ 0.56 ± 0.0315

vllm (pretrained=/root/autodl-tmp/80-256-4096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.578 ± 0.0221
strict-match 5 exact_match ↑ 0.574 ± 0.0221

vllm (pretrained=/root/autodl-tmp/80-512-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.564 ± 0.0314
strict-match 5 exact_match ↑ 0.564 ± 0.0314

vllm (pretrained=/root/autodl-tmp/80-512-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.570 ± 0.0222
strict-match 5 exact_match ↑ 0.566 ± 0.0222

vllm (pretrained=/root/autodl-tmp/80-512-8192,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups Version Filter n-shot Metric Value Stderr
mmlu 2 none acc ↑ 0.4246 ± 0.0167
- humanities 2 none acc ↑ 0.3897 ± 0.0344
- other 2 none acc ↑ 0.4667 ± 0.0356
- social sciences 2 none acc ↑ 0.4222 ± 0.0366
- stem 2 none acc ↑ 0.4211 ± 0.0290

vllm (pretrained=/root/autodl-tmp/81-512-8192,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.56 ± 0.0315
strict-match 5 exact_match ↑ 0.56 ± 0.0315

vllm (pretrained=/root/autodl-tmp/81-512-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.564 ± 0.0222
strict-match 5 exact_match ↑ 0.562 ± 0.0222

vllm (pretrained=/root/autodl-tmp/82-256-8192,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.564 ± 0.0314
strict-match 5 exact_match ↑ 0.564 ± 0.0314

vllm (pretrained=/root/autodl-tmp/82-256-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.586 ± 0.0220
strict-match 5 exact_match ↑ 0.580 ± 0.0221

vllm (pretrained=/root/autodl-tmp/82-256-8192,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups Version Filter n-shot Metric Value Stderr
mmlu 2 none acc ↑ 0.4292 ± 0.0166
- humanities 2 none acc ↑ 0.4051 ± 0.0340
- other 2 none acc ↑ 0.4718 ± 0.0355
- social sciences 2 none acc ↑ 0.4278 ± 0.0362
- stem 2 none acc ↑ 0.4175 ± 0.0289

vllm (pretrained=/root/autodl-tmp/82-512-8192,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.564 ± 0.0314
strict-match 5 exact_match ↑ 0.560 ± 0.0315

vllm (pretrained=/root/autodl-tmp/82-512-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.572 ± 0.0221
strict-match 5 exact_match ↑ 0.564 ± 0.0222

vllm (pretrained=/root/autodl-tmp/82-1024-8192,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.552 ± 0.0315
strict-match 5 exact_match ↑ 0.552 ± 0.0315

vllm (pretrained=/root/autodl-tmp/83-512-8192,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.532 ± 0.0316
strict-match 5 exact_match ↑ 0.532 ± 0.0316

vllm (pretrained=/root/autodl-tmp/84-256-8192,add_bos_token=true,max_model_len=8096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match ↑ 0.540 ± 0.0316
strict-match 5 exact_match ↑ 0.536 ± 0.0316

README history 1 version

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

  1. 2025-05-17Create README.md71b33de9.9 KB
    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.

Open in Abliteration