Huihui-Qwen3.8-27B-abliterated
huihui-ai Apache 2.0 out 2 months ago
27.8B parameters, images and text in, text out. 1 provider serves it.
Verified 2026-10-25 Rank #551 in LLMs · score 74.9 method v1 Sources 3 Downloads 30d 38.1k · #401 of 619 tracked
Changed this week: 2 new quantized builds on the Hub (92 in all).2026-10-24
Where to run it
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | Featherless | — | — | — | — | Oct 25 |
Where it's used
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; heavily run locally (92 quantized builds).
Will it fit?
Weights only, an estimate: leave room for the KV cache. Method
Fits a 64 GB Mac (unified memory) at INT4
About 17 GB at INT4, 33 GB at FP8, 67 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
What people built on it
112 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
112 derivatives on the Hub, most of them quantized
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Reference
Everything else, folded: the full tables, every source.
Use-case evidence
2 use cases
What it is good at 2 use cases
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Facts
14 fields
Listed in LLMs
Facts
- Huihui-Qwen3.8-27B-abliterated has 27.8B parameters. (safetensors metadata, observed —)
- License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed —)
- Created 2026-08-16, last modified 2026-08-24. (HF API, observed —)
| Pipeline tag | image-text-to-text |
|---|---|
| Library | transformers |
| Architectures | Qwen3_5ForConditionalGeneration |
| Model type | qwen3_5 |
| Params by dtype | BF16: 27,781,427,952 |
| Size band | L |
| MoE | no |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Hardware fit
3 estimates
Hardware fit ESTIMATE
Weights-only memory for Huihui-Qwen3.8-27B-abliterated: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 67 GB · Single high-end GPU (<80 GB) |
|---|---|
| FP8 (1 byte/param) | 33 GB · Single high-end GPU (<80 GB) |
| INT4 (0.5 bytes/param) | 17 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
All variants
40 variants
Lineage
Base model: Qwen/Qwen3.8-27B
quantized: 92finetune: 11merge: 3adapter: 6
Top variants (40 shown, ranked by downloads)
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| LessThanThreeAI/Qwen3.8-27B-Humanlike-Chat-GGUF | quantized | GGUF | — | — | apache-2.0 | LessThanThreeAI | 211 | 114.2k |
| sakamakismile/Huihui-Qwen3.8-27B-abliterated-NVFP4 | quantized | NVFP4 | 27.4B | L | apache-2.0 | sakamakismile | 66 | 102.4k |
| Barding-Defense/Qwen3.8-27B-huihui-abliterated-NVFP4-NInfer | quantized | NVFP4 | — | — | apache-2.0 | Barding-Defense | 12 | 69.3k |
| RentedNoodle/Qwen3.8-27B-GSQ-RCO-IQ3_XXS-Uncensored | quantized | GGUF | — | — | apache-2.0 | RentedNoodle | 67 | 51.8k |
| prithivMLmods/Qwen3.8-27B-abliterated-GGUF | quantized | GGUF | — | — | apache-2.0 | prithivMLmods | 7 | 42.1k |
| mradermacher/Huihui-Qwen3.8-27B-abliterated-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 5 | 39.2k |
| Barding-Defense/Qwen3.8-27B-huihui-abliterated-groupwise-int-NInfer | quantized | — | — | — | apache-2.0 | Barding-Defense | 1 | 37.8k |
| mradermacher/Huihui-Qwen3.8-27B-abliterated-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 18 | 28.8k |
| TelperionAI/Huihui-Qwen3.8-27B-abliterated-INT4-AWQ-GPTQ | quantized | INT4 | 7.84B | M | apache-2.0 | TelperionAI | 7 | 25.3k |
| WaveCut/Huihui-Qwen3.8-27B-abliterated-NInfer-v3 | quantized | — | — | — | apache-2.0 | WaveCut | 5 | 19.2k |
| lyf/Qwen3.8-27B-Huihui-Abliterated-NInfer-NVFP4 | quantized | NVFP4 | — | — | apache-2.0 | lyf | 19 | 15.3k |
| ArtomYuan/Qwen3.8-27B-abliterated-ROCmFPX | quantized | GGUF | — | — | apache-2.0 | ArtomYuan | 3 | 13.1k |
| edp1096/Huihui-RadixArk-Qwen3.8-27B-abliterated-NVFP4 | quantized | NVFP4 | 18.2B | L | — | edp1096 | 1 | 11.2k |
| gearwave00001/Huihui-Qwen3.8-27B-abliterated-nvfp4-NInfer | finetune | nvfp4 | — | — | apache-2.0 | gearwave00001 | 2 | 9.9k |
| munekazu/Huihui-Qwen3.8-27B-abliterated-FP8 | quantized | -FP8 | 27.8B | L | apache-2.0 | munekazu | 3 | 9.8k |
| lyf/Qwen3.8-27B-Huihui-Abliterated-NVFP4-MTP-VL | quantized | NVFP4 | 27.8B | L | apache-2.0 | lyf | 3 | 8.8k |
| dongfangshuo/Dongfangshuo-Qwen3.8-27B-NSFW | finetune | — | 27.8B | L | apache-2.0 | dongfangshuo | 10 | 7.8k |
| mitkox/Huihui-Qwen3.8-27B-abliterated-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | mitkox | 6 | 7.5k |
| AutisticAF/Huihui-Qwen3.8-27B-abliterated-mlx-4Bit | quantized | mlx | 26.9B | L | apache-2.0 | AutisticAF | 9 | 7.1k |
| renketong/Huihui-Qwen3.8-27B-abliterated-NVFP4-GGUF | quantized | NVFP4 | — | — | apache-2.0 | renketong | 1 | 7.1k |
| lued/Qwen3.8-27B-huihui-abliterated-INT8-W8A16-MTP | quantized | INT8 | 27.8B | L | apache-2.0 | lued | 4 | 6.9k |
| donghanasd/Huihui-Qwen3.8-27B-abliterated-KO-Ridge-3.7bpw-GGUF | quantized | 7bpw | — | — | apache-2.0 | donghanasd | 6 | 5.2k |
| IsValorum/Huihui-Qwen3.8-27B-Abliterated-VAL-APEX-I-NanoPlus-V3-GGUF | quantized | GGUF | — | — | apache-2.0 | IsValorum | 5 | 5.2k |
| IsValorum/Huihui-Qwen3.8-27B-Abliterated-VAL-APEX-I-MiniPlus-V3-GGUF | quantized | GGUF | — | — | apache-2.0 | IsValorum | 3 | 4.7k |
| ajaeger74/Huihui-Qwen3.8-27B-abliterated-Q6_K-GGUF | quantized | Q6_K | — | — | apache-2.0 | ajaeger74 | 2 | 4.4k |
| KostkaIT/Qwen3.8-27B-Huihui-Abliterated-oQ6e-MTP-MLX | quantized | MLX | 27.8B | L | apache-2.0 | KostkaIT | 6 | 3.9k |
| leoncca/Qwen3.8-27B-Huihui-AWQ | quantized | AWQ | 27.8B | L | apache-2.0 | leoncca | 1 | 4.1k |
| ailexleon/Huihui-Qwen3.8-27B-abliterated-mlx-4Bit | quantized | mlx | 27.4B | L | apache-2.0 | ailexleon | 1 | 3.7k |
| leoncca/Qwen3.8-27B-Huihui-Mixed-FP8 | quantized | -FP8 | 27.8B | L | apache-2.0 | leoncca | 1 | 3.5k |
| KostkaIT/Qwen3.8-27B-Huihui-Abliterated-oQ4e-MTP-MLX | quantized | MLX | 27.8B | L | apache-2.0 | KostkaIT | 4 | 3.3k |
| ailexleon/Huihui-Qwen3.8-27B-abliterated-mlx-6Bit | quantized | mlx | 27.4B | L | apache-2.0 | ailexleon | 1 | 3.2k |
| batsclamp/Huihui-Qwen3.8-27B-abliterated-FP8-v2 | quantized | -FP8 | 27.4B | L | apache-2.0 | batsclamp | 0 | 3.1k |
| ailexleon/Huihui-Qwen3.8-27B-abliterated-mlx-8Bit | quantized | mlx | 27.4B | L | apache-2.0 | ailexleon | 1 | 2.9k |
| ababaka/Huihui-Qwen3.8-27B-Abliterated-W4A16-AutoRound | quantized | -W4A16 | 6.26B | M | apache-2.0 | ababaka | 3 | 2.7k |
| Guile/Qwen3.8-27B-Humanlike-Chat-GGUF | quantized | GGUF | — | — | apache-2.0 | Guile | 6 | 2.5k |
| trailio/Huihui-Qwen3.8-27B-abliterated-NVFP8 | quantized | — | 27.8B | L | apache-2.0 | trailio | 0 | 2.5k |
| augustine223/Huihui-Qwen3.8-27B-abliterated-KO-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | augustine223 | 1 | 2.4k |
| dongfangshuo/Dongfangshuo-Qwen3.8-27B-NSFW-v4 | finetune | GGUF | 26.9B | L | other | dongfangshuo | 6 | 2.1k |
| satellitedown/Huihui-Qwen3.8-27B-abliterated-NVFP4-NInfer-v3 | quantized | NVFP4 | — | — | apache-2.0 | satellitedown | 1 | 2.3k |
| Yirasumi/Huihui-Qwen3.8-27B-abliterated-INT4-W4A16 | quantized | INT4 | 5.16B | M | apache-2.0 | Yirasumi | 0 | 2.1k |
Repository changelog
1 change
All providers
1 provider · 1 endpoint
Where to run 1 providers · 1 endpoints
Hosted endpoints observed for Huihui-Qwen3.8-27B-abliterated, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| Featherless hf | huihui-ai/Huihui-Qwen3.8-27B-abliterated | — | — | — | — | — | — | — | live | 2026-10-25 |
Similar models
3 models
Similar models
- Qwen3.8-27B · 27.8B same size band (L), same task (image-text-to-text), more providers, permissive license
- Qwen3.6-27B · 27.8B same size band (L), same task (image-text-to-text), more providers, permissive license
- Qwen3.5-27B · 27.8B same size band (L), same task (image-text-to-text), more providers, permissive license
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-09 02:04 UTC · Last verified 2026-10-25
Follow Huihui-Qwen3.8-27B-abliterated
Watch this model
- /feeds/models/huihui-ai--Huihui-Qwen3.8-27B-abliterated.xml(RSS 2.0) · every change to Huihui-Qwen3.8-27B-abliterated
- /feeds/families/huihui-ai--Huihui.xml(RSS 2.0) · every change across the Huihui family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.