# Models, Page 166 of 525
Source: https://savrn.com/models/page/166
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

SAVRN Model Hub

# Open-Weight Models

An open-weight model is an AI model whose trained weights are published for anyone to download. The weights are what the model learned in training. With a copy of them you can run the model on hardware you control and train it further on your own data.

Open weights are not the same as open source. Many publishers release the weights without the training data or code, and the license sets what you may do with the model. This library puts each model's full card, architecture, files, license and published evaluations on one page.

12,579Models

3,583Datasets

291Papers

7,578Publishers

12,850Sourced relationships

Updated 2026-10-23 · [How the library is built](https://savrn.com/model-hub/methodology)

## Filter Models

Task

- [Text generation1,548](https://savrn.com/models?task=Text+generation)
- [Image and text to text577](https://savrn.com/models?task=Image+and+text+to+text)
- [Feature extraction398](https://savrn.com/models?task=Feature+extraction)
- [Robotics398](https://savrn.com/models?task=Robotics)
- [Text classification331](https://savrn.com/models?task=Text+classification)
- [Text to image232](https://savrn.com/models?task=Text+to+image)
- [Speech recognition187](https://savrn.com/models?task=Speech+recognition)
- [Text to speech145](https://savrn.com/models?task=Text+to+speech)
- [Image classification139](https://savrn.com/models?task=Image+classification)
- [Audio classification130](https://savrn.com/models?task=Audio+classification)
- [Sentence similarity103](https://savrn.com/models?task=Sentence+similarity)
- [Reinforcement learning100](https://savrn.com/models?task=Reinforcement+learning)
- [Token classification100](https://savrn.com/models?task=Token+classification)
- [Zero-shot classification94](https://savrn.com/models?task=Zero-shot+classification)
- [Image segmentation90](https://savrn.com/models?task=Image+segmentation)
- [Object detection86](https://savrn.com/models?task=Object+detection)
- [Translation80](https://savrn.com/models?task=Translation)
- [Any to any79](https://savrn.com/models?task=Any+to+any)
- [Question answering73](https://savrn.com/models?task=Question+answering)
- [Fill mask69](https://savrn.com/models?task=Fill+mask)
- [Time series forecasting64](https://savrn.com/models?task=Time+series+forecasting)
- [Video classification61](https://savrn.com/models?task=Video+classification)
- [Image to text60](https://savrn.com/models?task=Image+to+text)
- [Text to video55](https://savrn.com/models?task=Text+to+video)

Library

- [transformers3,001](https://savrn.com/models?library=transformers)
- [peft367](https://savrn.com/models?library=peft)
- [diffusers206](https://savrn.com/models?library=diffusers)
- [pytorch198](https://savrn.com/models?library=pytorch)
- [gguf187](https://savrn.com/models?library=gguf)
- [mlx169](https://savrn.com/models?library=mlx)
- [lerobot152](https://savrn.com/models?library=lerobot)
- [sentence-transformers139](https://savrn.com/models?library=sentence-transformers)
- [timm49](https://savrn.com/models?library=timm)
- [llama.cpp46](https://savrn.com/models?library=llama.cpp)
- [transformers.js45](https://savrn.com/models?library=transformers.js)
- [flaxspeaker44](https://savrn.com/models?library=flaxspeaker)
- [diffusion-single-file40](https://savrn.com/models?library=diffusion-single-file)
- [onnx40](https://savrn.com/models?library=onnx)
- [vllm40](https://savrn.com/models?library=vllm)
- [openpi35](https://savrn.com/models?library=openpi)
- [minimax-h328](https://savrn.com/models?library=minimax-h3)
- [ultralytics24](https://savrn.com/models?library=ultralytics)
- [PaddleOCR22](https://savrn.com/models?library=PaddleOCR)
- [coreml22](https://savrn.com/models?library=coreml)
- [exllamav322](https://savrn.com/models?library=exllamav3)
- [openvino20](https://savrn.com/models?library=openvino)
- [litert17](https://savrn.com/models?library=litert)
- [litert-lm16](https://savrn.com/models?library=litert-lm)

Modality

- [Other7,212](https://savrn.com/models?modality=Other)
- [Text2,838](https://savrn.com/models?modality=Text)
- [Image and text637](https://savrn.com/models?modality=Image+and+text)
- [Image598](https://savrn.com/models?modality=Image)
- [Control498](https://savrn.com/models?modality=Control)
- [Audio481](https://savrn.com/models?modality=Audio)
- [Video116](https://savrn.com/models?modality=Video)
- [Multimodal79](https://savrn.com/models?modality=Multimodal)
- [Time series64](https://savrn.com/models?modality=Time+series)
- [Tabular56](https://savrn.com/models?modality=Tabular)

License

- [apache-2.03,344](https://savrn.com/models?license=apache-2.0)
- [mit1,319](https://savrn.com/models?license=mit)
- [other796](https://savrn.com/models?license=other)
- [cc-by-nc-4.0213](https://savrn.com/models?license=cc-by-nc-4.0)
- [cc-by-4.0164](https://savrn.com/models?license=cc-by-4.0)
- [bsd-3-clause108](https://savrn.com/models?license=bsd-3-clause)
- [gemma63](https://savrn.com/models?license=gemma)
- [cc-by-nc-sa-4.055](https://savrn.com/models?license=cc-by-nc-sa-4.0)
- [llama3.240](https://savrn.com/models?license=llama3.2)
- [unknown33](https://savrn.com/models?license=unknown)
- [agpl-3.032](https://savrn.com/models?license=agpl-3.0)
- [llama3.128](https://savrn.com/models?license=llama3.1)
- [cc-by-sa-4.020](https://savrn.com/models?license=cc-by-sa-4.0)
- [gpl-3.020](https://savrn.com/models?license=gpl-3.0)
- [openrail++17](https://savrn.com/models?license=openrail%2B%2B)
- [creativeml-openrail-m15](https://savrn.com/models?license=creativeml-openrail-m)
- [openmdw-1.115](https://savrn.com/models?license=openmdw-1.1)
- [cc-by-nc-nd-4.014](https://savrn.com/models?license=cc-by-nc-nd-4.0)
- [llama314](https://savrn.com/models?license=llama3)
- [openrail14](https://savrn.com/models?license=openrail)
- [llama28](https://savrn.com/models?license=llama2)
- [afl-3.07](https://savrn.com/models?license=afl-3.0)
- [osl-3.06](https://savrn.com/models?license=osl-3.0)
- [cc0-1.04](https://savrn.com/models?license=cc0-1.0)

Access

- [Open weights12,153](https://savrn.com/models?access=open_access)
- [Access requested at publisher426](https://savrn.com/models?access=gated)

Publisher

- [Team Mradermacher328](https://savrn.com/models?publisher=mradermacher)
- [Yurun Yuan115](https://savrn.com/models?publisher=ryanyr)
- [RunningHub.ai105](https://savrn.com/models?publisher=runninghubai)
- [Qwen91](https://savrn.com/models?publisher=qwen)
- [Jenny Wilson85](https://savrn.com/models?publisher=jenny08311)
- [Jessen Pu77](https://savrn.com/models?publisher=putty77)
- [RKNNAI73](https://savrn.com/models?publisher=rknnai)
- [AI at Meta72](https://savrn.com/models?publisher=facebook)
- [Convergent Intelligence66](https://savrn.com/models?publisher=reaperdoesntknow)
- [Yurun Yuan65](https://savrn.com/models?publisher=yurunyyr)
- [Google64](https://savrn.com/models?publisher=google)
- [Yurun Yuan62](https://savrn.com/models?publisher=yuruny)
- [Firza Hadzami61](https://savrn.com/models?publisher=firzahdzm)
- [Zeyuan Yang59](https://savrn.com/models?publisher=miiche)
- [Yurun Yuan58](https://savrn.com/models?publisher=yyuan244)
- [Yanggangu57](https://savrn.com/models?publisher=yanggangu)
- [Hongao56](https://savrn.com/models?publisher=cisco1963)
- [Christopher Subia-Waud55](https://savrn.com/models?publisher=cwaud)
- [Francesca Padovani53](https://savrn.com/models?publisher=francesca9805)
- [Khazzz1c53](https://savrn.com/models?publisher=khazic)
- [Alfred Smith52](https://savrn.com/models?publisher=alfred8995)
- [Szekely52](https://savrn.com/models?publisher=attila115)
- [NVIDIA50](https://savrn.com/models?publisher=nvidia)
- [Wang49](https://savrn.com/models?publisher=davidwdw)

[Most downloaded](https://savrn.com/models?sort=downloads)[Most liked](https://savrn.com/models?sort=likes)[Recently updated](https://savrn.com/models?sort=updated)[Name](https://savrn.com/models?sort=name)[Largest](https://savrn.com/models?sort=params)

12,579 models, sorted by most downloaded.

Model · Image and text to text

### [Huihui-GLM-5.3-Flash-abliterated-GGUF](https://savrn.com/models/huihui-glm-5-3-flash-abliterated-gguf-2)

[Petyxavich](https://savrn.com/model-publishers/petyx123)

This is an uncensored version of zai-org/GLM-5.3-Flash created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model. The latest version of llama.cpp now officially supports zai-org/GLM-5.3-Flash. We have updated the first file of each quantized version, so you only need to download the first file. GGUFs come from unsloth/GLM-5.3-Flash-GGUF. Only layers 15 to 35 (0-based indexing) have been ablated, while the other layers and all expert modules remain unablated. To run, please unslothai llama.cpp PR, - Risk of Sensitive or Controversial Outputs: This model’s safety filtering has…

Open weights mit transformers

[View model](https://savrn.com/models/huihui-glm-5-3-flash-abliterated-gguf-2)

Model · Text generation

### [Huihui-Kolibri-1-abliterated-GGUF](https://savrn.com/models/huihui-kolibri-1-abliterated-gguf-2)

[Okardu](https://savrn.com/model-publishers/okardu)

This is an uncensored version of Aleph-Alpha/Kolibri-1 created with abliteration. This ablation was performed entirely using llama.cpp's GGUF model, without relying on Transformers, For the specific ablation method, please refer to cvector-generator. GGUFs come from Hob-forge/Kolibri-1-GGUF. The ablation did not modify all expert modules. Stock llama.cpp does not support the kolibri1 architecture yet. Apply kolibri1-llama.cpp.patch (included in this repo) to llama.cpp at upstream commit 836d571, then build: - Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content.…

Open weights apache-2.0 vllm

[View model](https://savrn.com/models/huihui-kolibri-1-abliterated-gguf-2)

Model · Text generation

### [Huihui-NeoHorse-1-4B-abliterated-GGUF](https://savrn.com/models/huihui-neohorse-1-4b-abliterated-gguf)

[Huihui.ai](https://savrn.com/model-publishers/huihui-ai)

This is an uncensored version of TokenRhythm/NeoHorse-1-4B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. Layers 5-17 are being ablated (0-based indexing). The MTP and Visual components were extracted from the original Qwen/Qwen3.5-4B and can provide excellent support. Use the latest llama.cpp, - Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated…

Open weights apache-2.0 transformers

[View model](https://savrn.com/models/huihui-neohorse-1-4b-abliterated-gguf)

Model · Image and text to text

### [Huihui-Qwen3.8-27B-abliterated-FP8](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-fp8)

[Guang Yang](https://savrn.com/model-publishers/gyang274)

An FP8 quantization of huihui-ai/Huihui-Qwen3.8-27B-abliterated that reproduces Qwen's official FP8 recipe byte for byte on every tensor the abliteration did not touch. The weights are block-wise FP8 E4M3 with 128×128 scales, in the same layout as Qwen/Qwen3.8-27B-FP8. The MTP head and vision tower are preserved, and the checkpoint is 28.75 GiB. It loads through the same code path as the official FP8 checkpoint. This model was built and validated as part of a single-GPU deployment study. The full notes, scripts and raw data are at github.com/gyang274/ms-Qwen3.8-27B. The quantizer (quantizefp8like.py) uses the official FP8 checkpoint as a layout template. It quantizes exactly the tensors…

Open weights apache-2.0 27.8B parameters 262,144 tokens transformers

[View model](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-fp8)

Model · Image and text to text

### [Huihui-Qwen3.8-27B-abliterated-GGUF](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-gguf-2)

[Marabas Valdegot](https://savrn.com/model-publishers/leemantras)

This is an uncensored version of Qwen/Qwen3.8-27B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. The newly added Huihui-Qwen3.8-27B-abliterated-Ternary series come from prism-ml/Ternary-Bonsai-2-27B-gguf have been ablated, while the other layers remain unablated. It may come with a small disclaimer warning. The size after conversion may differ from the original GGUF (Some of the weights are converted from PTQ1 to Q2K or Q3K.). This is just a test/validation. The ternary hybrid-attention kernels live in the PrismML-Eng/llama.cpp fork.…

Open weights apache-2.0 transformers

[View model](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-gguf-2)

Model · Image and text to text

### [Huihui-Qwen3.8-27B-abliterated-GGUF](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-gguf-3)

[Ahmed](https://savrn.com/model-publishers/ahmed-amedo)

This is an uncensored version of Qwen/Qwen3.8-27B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. The newly added Huihui-Qwen3.8-27B-abliterated-Underdog-Saluki series come from ConwayResearch/Underdog-Saluki-27B-1.0 and ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF. Only layers 22 to 52 (0-based indexing) have been ablated, while the other layers remain unablated. It may come with a small disclaimer warning. Due to ConwayResearch/Underdog-Saluki-27B-1.0 not providing an imatrix file, we will attempt to perform ablation by mixing the GGUF…

Open weights apache-2.0 transformers

[View model](https://savrn.com/models/huihui-qwen3-8-27b-abliterated-gguf-3)

Model

### [Huihui-Qwen3.8-Flash-Next-abliterated-GGUF](https://savrn.com/models/huihui-qwen3-8-flash-next-abliterated-gguf)

[Team Mradermacher](https://savrn.com/model-publishers/mradermacher)

static quants of https://huggingface.co/huihui-ai/Huihui-Qwen3.8-Flash-Next-abliterated For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some…

Open weights other transformers

[View model](https://savrn.com/models/huihui-qwen3-8-flash-next-abliterated-gguf)

Model · Image and text to text

### [Huihui-Qwen3.8-Flash-Next-abliterated-v2-GGUF](https://savrn.com/models/huihui-qwen3-8-flash-next-abliterated-v2-gguf-2)

[Benjamin Kauffman](https://savrn.com/model-publishers/benmonkey11)

This is an uncensored version of Qwen/Qwen3.8-Flash-Next created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. This version Compared to the first version, the ablation emphasis has been adjusted, and none of the experts have been ablated. Suitable for novel writing and programming. The newly added Swift series come from ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF. The unsloth series come from unsloth/Qwen3.8-Flash-Next-GGUF. This unsloth and Swift series supports the latest versions of both Strata and llama.cpp. Strata has byte-level and SHA256…

Open weights other transformers

[View model](https://savrn.com/models/huihui-qwen3-8-flash-next-abliterated-v2-gguf-2)

Model · Any to any

### [Huihui-gemma-4-12B-it-abliterated](https://savrn.com/models/huihui-gemma-4-12b-it-abliterated)

[Isaac Jaxxon](https://savrn.com/model-publishers/timelesssoul)

This is an uncensored version of google/gemma-4-12B-it created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. Note For this model, both the thinking mode and the non-thinking mode have been completely abliterated. Only layers 23-28 have been abliterated. Important Notice: The weights of the earliest released google/gemma-4-12B-it model have issues. We have already re-ablated and re-uploaded the model. If you have already downloaded it, please re-download. Please use the latest version of ollama You can use huihuiai/gemma-4-abliterated:12b…

Open weights apache-2.0 12B parameters 131,072 tokens transformers

[View model](https://savrn.com/models/huihui-gemma-4-12b-it-abliterated)

Model · Any to any

### [Huihui-gemma-4-12B-it-abliterated](https://savrn.com/models/huihui-gemma-4-12b-it-abliterated-2)

[Daniel](https://savrn.com/model-publishers/ibyteohdear)

This is an uncensored version of google/gemma-4-12B-it created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. Note For this model, both the thinking mode and the non-thinking mode have been completely abliterated. Only layers 23-28 have been abliterated. Important Notice: The weights of the earliest released google/gemma-4-12B-it model have issues. We have already re-ablated and re-uploaded the model. If you have already downloaded it, please re-download. Please use the latest version of ollama You can use huihuiai/gemma-4-abliterated:12b…

Open weights apache-2.0 transformers

[View model](https://savrn.com/models/huihui-gemma-4-12b-it-abliterated-2)

Model

### [Humanizer-12B-Draftfit-DFlash](https://savrn.com/models/humanizer-12b-draftfit-dflash)

[JayYun](https://savrn.com/model-publishers/jayyun98)

A validation checkpoint for Draftfit — train DFlash-family drafts jointly, convert compatible families, and continue training existing drafts. This DFlash draft accelerates Humanizer; it requires the target model. Target / teacher: Humanizer. Initial draft weights: Gemma4 DFlash. This is a target-specific speculative draft, not a standalone Humanizer fine-tune. 4/8-bit target with FP32 floating computation · 4-bit drafts. Median throughput includes prefill and verification; acceptance is accepted/proposed draft tokens. Same 10 held-out inputs ×2 repeats per model, greedy, up to 64 output tokens, draft cap 3, lookup off. All 160 requests matched target-only tokens, including…

Open weights apache-2.0 727M parameters 262,144 tokens

[View model](https://savrn.com/models/humanizer-12b-draftfit-dflash)

Model

### [Humanizer-12B-Draftfit-DFlash2](https://savrn.com/models/humanizer-12b-draftfit-dflash2)

[JayYun](https://savrn.com/model-publishers/jayyun98)

A validation checkpoint for Draftfit — train DFlash-family drafts jointly, convert compatible families, and continue training existing drafts. This DFlash2 draft accelerates Humanizer; it requires the target model. Target / teacher: Humanizer. Initial draft weights: Gemma4 DFlash. This is a target-specific speculative draft, not a standalone Humanizer fine-tune. 4/8-bit target with FP32 floating computation · 4-bit drafts. Median throughput includes prefill and verification; acceptance is accepted/proposed draft tokens. Same 10 held-out inputs ×2 repeats per model, greedy, up to 64 output tokens, draft cap 3, lookup off. All 160 requests matched target-only tokens, including…

Open weights apache-2.0 900M parameters 262,144 tokens

[View model](https://savrn.com/models/humanizer-12b-draftfit-dflash2)

Model

### [Humanizer-12B-Draftfit-DSpark](https://savrn.com/models/humanizer-12b-draftfit-dspark)

[JayYun](https://savrn.com/model-publishers/jayyun98)

A validation checkpoint for Draftfit — train DFlash-family drafts jointly, convert compatible families, and continue training existing drafts. This DSpark draft accelerates Humanizer; it requires the target model. Target / teacher: Humanizer. Initial draft weights: Gemma4 DFlash. This is a target-specific speculative draft, not a standalone Humanizer fine-tune. 4/8-bit target with FP32 floating computation · 4-bit drafts. Median throughput includes prefill and verification; acceptance is accepted/proposed draft tokens. Same 10 held-out inputs ×2 repeats per model, greedy, up to 64 output tokens, draft cap 3, lookup off. All 160 requests matched target-only tokens, including…

Open weights apache-2.0 862M parameters 262,144 tokens

[View model](https://savrn.com/models/humanizer-12b-draftfit-dspark)

Model · Text generation

### [Hunter-4B](https://savrn.com/models/hunter-4b)

[Lalakai Labs](https://savrn.com/model-publishers/lalakai)

Introducing Hunter 4B by Lalakai Labs. A 4B-parameter conversational reasoning model, distributed as GGUF weights for fast local inference with llama.cpp, Ollama, LM Studio, and any other GGUF-compatible runtime. GGUF is a safe, mmap-friendly format — no pickle files here. Hunter reasons out loud: it works through problems step by step inside... traces before giving you the final answer — handy for math, code, and multi-step tasks. Despite its small footprint, Hunter-4B is built to rival the performance of models in the 10–15B parameter range on reasoning, coding, and cyber tasks. Hunter 4B is a strong fine-tune of the Qwen 3.5 4B, trained on ~180,000 high-density examples. No raw web…

Open weights mit

[View model](https://savrn.com/models/hunter-4b)

Model

### [Hush-Nano-22M](https://savrn.com/models/hush-nano-22m)

[Leecz](https://savrn.com/model-publishers/soulitude)

Open weights apache-2.0 23M parameters 1,024 tokens

[View model](https://savrn.com/models/hush-nano-22m)

Model · Text generation

### [Hush-Nano-Chat](https://savrn.com/models/hush-nano-chat)

[Leecz](https://savrn.com/model-publishers/soulitude)

Hush-Nano-Chat is an English, single-turn instruction-tuned version of Soulitude/Hush-Nano. It starts from the 22M-parameter pretrained model and uses supervised fine-tuning (SFT) on instruction–response pairs. Due to the model's limited parameters, its response can be inaccurate, incomplete, or inconsistent. Hush-Nano-Chat has the following features: The base model was pretrained on 8.5B tokens (8,554,042,292) drawn from the following subsets: Then it was fine-tuned on a mixture of the following datasets: unsloth/alpaca-cleaned, databricks/databricks-dolly-15k, and HuggingFaceH4/norobots. Only the assistant response and ending EOS token contribute to the training loss. Zero-shot normalized…

Open weights apache-2.0 23M parameters 1,024 tokens transformers

[View model](https://savrn.com/models/hush-nano-chat)

Model

### [Hy-MT2-1.8B-GGUF](https://savrn.com/models/hy-mt2-1-8b-gguf)

[ExNihilus](https://savrn.com/model-publishers/exnihilus)

Unmodified copy of selected files from tencent/Hy-MT2-1.8B-GGUF at revision a0c709d9fac510f2c807aa3af52872340dc37a4a, kept so that SubVox Pro (Unity editor plugin by ExNihilus) can download a stable, pinned version of the model. All credit goes to the original authors: the conversion by tencent and the original model tencent/Hy-MT2-1.8B. - Hy-MT2-1.8B-Q4KM.gguf

Open weights apache-2.0

[View model](https://savrn.com/models/hy-mt2-1-8b-gguf)

Model

### [Hy-MT2-30B-A3B-oQ6e-fp16](https://savrn.com/models/hy-mt2-30b-a3b-oq6e-fp16)

[Vanchye](https://savrn.com/model-publishers/vanch007)

Open weights

[View model](https://savrn.com/models/hy-mt2-30b-a3b-oq6e-fp16)

Model · Text generation

### [Hy-Memory-30B-A3B-GGUF](https://savrn.com/models/hy-memory-30b-a3b-gguf)

[Ben](https://savrn.com/model-publishers/chiefbroseph)

An Asymmetric Hybrid Q4KM (Q80 / Q6K / Q4KM) GGUF quantization of tencent/Hy-Memory-30B-A3B, calibrated with an importance matrix (imatrix) and tuned specifically for single 24 GB GPUs (RTX 3090 / RTX 4090) as well as zero-VRAM CPU sidecar deployments. Hy-Memory-30B-A3B (HYV3ForCausalLM / hyv3 architecture) is a sparse Mixture-of-Experts memory writer model with 30.66B total parameters and ~3.0B active parameters per token (39 MoE layers, 128 routed experts, top-8 routing + 1 shared expert, and 8:1 GQA). Because 94.5% of the model's parameters (28.97B) live inside the 128 sparse routed experts while only 5.5% (1.69B) make up the always-active backbone, uniform quantization leaves massive…

Open weights other gguf

[View model](https://savrn.com/models/hy-memory-30b-a3b-gguf)

Model · Text generation

### [Hy3-Razor-154B-A18B-E96of192](https://savrn.com/models/hy3-razor-154b-a18b-e96of192)

[Mingyang Song](https://savrn.com/model-publishers/nickyang)

Hy3 with half of its routed experts removed by RAZOR, a training-free expert pruning method. Every MoE layer keeps 96 of its original 192 routed experts. No gradient updates or recovery training were applied: the retained weights are the base model's own weights. Pruning touches only the routed expert pool. Attention, shared experts, the embedding and the LM head are untouched, so the compute per token drops only by the share of expert FLOPs that the removed experts would have contributed. The other budget is Requires a Transformers build containing the native hyv3 implementation. Weights are bfloat16. RAZOR asks whether the surviving computation can replace an expert's function, rather…

Open weights apache-2.0 153.8B parameters 262,144 tokens transformers

[View model](https://savrn.com/models/hy3-razor-154b-a18b-e96of192)

Model · Text generation

### [Hy3-Razor-226B-A18B-E144of192](https://savrn.com/models/hy3-razor-226b-a18b-e144of192)

[Mingyang Song](https://savrn.com/model-publishers/nickyang)

Hy3 with a quarter of its routed experts removed by RAZOR, a training-free expert pruning method. Every MoE layer keeps 144 of its original 192 routed experts. No gradient updates or recovery training were applied: the retained weights are the base model's own weights. Pruning touches only the routed expert pool. Attention, shared experts, the embedding and the LM head are untouched, so the compute per token drops only by the share of expert FLOPs that the removed experts would have contributed. The other budget is Requires a Transformers build containing the native hyv3 implementation. Weights are bfloat16. RAZOR asks whether the surviving computation can replace an expert's function…

Open weights apache-2.0 226.3B parameters 262,144 tokens transformers

[View model](https://savrn.com/models/hy3-razor-226b-a18b-e144of192)

Model

### [Hy3-nvidia-FlagOS](https://savrn.com/models/hy3-nvidia-flagos)

[FlagRelease](https://savrn.com/model-publishers/flagrelease)

Hy3 is Tencent Hunyuan team's next-generation MoE large model with integrated fast and slow thinking: 295B total parameters, 21B active parameters (plus 3.8B MTP layer parameters), 192 experts with top-8 activation, supporting 256K context. Compared to the Preview version released in late April, Hy3 has achieved a comprehensive leap in intelligence through incorporating real-world business feedback, scaling up RL compute, and improving post-training data quality, significantly outperforming open-source models of similar size. - Out-of-the-box inference scripts with pre-configured hardware and software parameters - Released FlagOS-Nvidia container image supporting deployment within minutes…

Open weights apache-2.0 262,144 tokens

[View model](https://savrn.com/models/hy3-nvidia-flagos)

Model · Text to image

### [Hybrid-Sensitivity-Weighted-Quantization-SDXL-ConvRot-INT8](https://savrn.com/models/hybrid-sensitivity-weighted-quantization-sdxl-convrot-int8)

[GJL](https://savrn.com/model-publishers/ussoewwin)

High-fidelity ConvRot INT8 reverse hybrid quantization for SDXL diffusion models. HSWQ uses per-layer trajectory-impact measurement instead of naive uniform cast, converting only the K lowest-impact layers to ConvRot INT8 while keeping every other layer at FP16. This is highly useful for users who need to strictly manage their VRAM resources while maintaining maximum image quality. Reverse hybrid (diag → reverse): The FP16 checkpoint is the only input. A per-layer trajectory-impact measurement (sdxl/diagimpactsdxl.py) injects each candidate layer's ConvRot INT8 reconstruction into the FP16 model one at a time and runs the production sampler — recording the final-latent drift. The K…

Open weights other nunchaku

[View model](https://savrn.com/models/hybrid-sensitivity-weighted-quantization-sdxl-convrot-int8)

Model · Text generation

### [HybridAgent-14B-Agentic-imatrix-GGUF](https://savrn.com/models/hybridagent-14b-agentic-imatrix-gguf)

[Liodon AI](https://savrn.com/model-publishers/liodon-ai)

Open weights other gguf

[View model](https://savrn.com/models/hybridagent-14b-agentic-imatrix-gguf)

## Model Collections

Hand-picked starting points, each with the reason it exists.

Collection · 4 entries

### [Embedding models for retrieval](https://savrn.com/collections/embedding-models-for-retrieval)

Sentence and document embedding models used to build retrieval systems. Dimension and sequence length matter more than size here, and both come from the publisher.

[View collection](https://savrn.com/collections/embedding-models-for-retrieval)

Collection · 4 entries

### [Models that fit on one accelerator](https://savrn.com/collections/models-that-run-on-one-accelerator)

Models whose publisher-reported parameter count puts them within reach of a single accelerator at common precisions. Memory needed depends on precision and serving configuration, so treat the parameter count as the starting point, not the answer.

[View collection](https://savrn.com/collections/models-that-run-on-one-accelerator)

Collection · 6 entries

### [Open-weight text models worth knowing](https://savrn.com/collections/open-weight-text-models-worth-knowing)

Widely used open-weight language models, chosen because each one is a distinct family rather than a variant of the one above it. Selection, not a ranking.

[View collection](https://savrn.com/collections/open-weight-text-models-worth-knowing)

Collection · 3 entries

### [Speech and audio models](https://savrn.com/collections/speech-and-audio)

Recognition and synthesis models, grouped so the two directions are easy to compare.

[View collection](https://savrn.com/collections/speech-and-audio)

## Open-Weight Models Explained

### What is an open-weight model?

An AI model whose trained weights are published for anyone to download, so it can be run, tested and fine-tuned on hardware the user controls.

### Is an open-weight model the same as open source?

Not always. Open weights means the trained model can be downloaded. Open source usually also means the training code and data are available and the license allows broad reuse. Many open-weight models release the weights only.

### Can I use an open-weight model commercially?

It depends on the license. Apache 2.0 and MIT allow commercial use. Other licenses limit it, for example to non-commercial use or below a set number of users. Every model page here shows its license.

### How much memory does an open-weight model need?

About two bytes per parameter at 16-bit precision, so a 7-billion-parameter model needs roughly 14 GB for its weights, plus memory for the context it processes. Each model page lists its parameter count and the size of its files.

## Related SAVRN Research

The hub sits beside SAVRN's market data and infrastructure research: what models cost to run, and what it takes to run them.

SAVRN Index

### [What open models cost to run](https://savrn.com/ai-index/pricing/open-models)

The same open-weight model priced by every host that serves it, per million tokens.

Research Hub

### [Data center trackers and maps](https://savrn.com/data-center-research)

Moratoriums, permits, power, water and capital behind the facilities that run these models.

Method

### [How the Model Hub is built](https://savrn.com/model-hub/methodology)

Sources, evidence labels, refresh behaviour, and the limits of every comparison here.