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.
Updated 2026-10-23 · How the library is built
12,579 models, sorted by most downloaded.
Mephisto-1.5-4B is a 4B-parameter agentic language model constructed via multi-stage model merging using mergekit. The model integrates reasoning, coding, and agentic capabilities into a single 4B checkpoint based on the Qwen3.5-4B architecture, targeting performance exceeding Mephisto-4B-v2 on Japanese and agentic benchmarks while maintaining deployment feasibility on consumer hardware (RTX 3060 12GB). The model is constructed through a three-stage merging pipeline: Normalized spherical linear interpolation (NuSLERP) on task vectors derived from a common base model (Qwen3.5-4B). This stage fuses a reasoning-distilled model (Opus-reasoning distillation) with a code-specialized model.…
ComfyUI-ready INT8 ConvRot build of Viggle/Meridian for MiniMax H3. Build preparado para ComfyUI do Viggle/Meridian, um modelo de recamera baseado no MiniMax H3. MeridianH3Cameraint8pruned.safetensors Place it in / Coloque em: meridiandmdloracomfyui.safetensors Place it in / Coloque em: Restart ComfyUI after installing the files. Reinicie o ComfyUI depois de instalar os arquivos. Meridian generates a new camera view from a source video using a camera warp video as guidance. O Meridian recria o vídeo a partir de uma nova câmera usando um vídeo de warp da câmera como guia. The warp can be generated with: O warp pode ser gerado com: The camera motion comes from the warp video. O movimento de…
static quants of https://huggingface.co/Hikari07jp/MiMo-V2.6-Distill-Qwen-9B-Ablitrated
EXL3 4.0 bpw conversion of XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B. Validated locally on an NVIDIA GeForce RTX 4090. The throughput figures are short local smoke measurements, not standardized cross-system benchmarks. Tool-call validation produced a structured getweather call for Toronto. Vision validation used a generated test image containing a large red square; the model correctly returned red. The source weights were not modified. Two local metadata normalizations were required for the current ExLlamaV3 conversion path: 1. The Qwen3.5 processor metadata declared Qwen2VLImageProcessor; the local conversion copy was normalized to Qwen2VLImageProcessorFast to match the ExLlamaV3 Qwen3.5…
Patched with the MTP draft tensors and the recommended amended jinja template from the ggml-org model page. ggml-org - Model unsloth - MTP donor model Coder543 - Jinja fix buzz - Grafting script Qwen-chan - Doing the actual work
Training in progress. Weights expected within the next 2 weeks. You can monitor progress here: https://training.armand0e.com Apologies for the delay, had some issues in grading. The training run was rolled back to before the bug was introduced. New ETA is ~5 Days
static quants of https://huggingface.co/Hikari07jp/MiMo-V2.6-Distill-Qwen-9B-Ablitrated 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…
weighted/imatrix quants of https://huggingface.co/Hikari07jp/MiMo-V2.6-Distill-Qwen-9B-ablit
Scaling Reinforcement Learning Toward Self-Improvement MiMo-V2.6-Flash-RL is the efficiency-balanced checkpoint of the MiMo-V2.6 series. The series is built to scale reinforcement learning toward self-improvement — scaling RL compute, environment diversity, and grader compute together, so the model keeps expanding its capability frontier through exploration and feedback. Key features include: - Native Omnimodal + Long Horizon: Text, image, video, and audio in one model; 1M tokens for long repositories, tool traces, and multi-session agent runs. - Multi-Prefix Multi-Teacher On-Policy Distillation (MOPD2): After mixed RL, MOPD2 combines autonomous student rollouts with prefix-conditioned…
Scaling Reinforcement Learning Toward Self-Improvement MiMo-V2.6-Flash-RL is the efficiency-balanced checkpoint of the MiMo-V2.6 series. The series is built to scale reinforcement learning toward self-improvement — scaling RL compute, environment diversity, and grader compute together, so the model keeps expanding its capability frontier through exploration and feedback. Key features include: - Native Omnimodal + Long Horizon: Text, image, video, and audio in one model; 1M tokens for long repositories, tool traces, and multi-session agent runs. - Multi-Prefix Multi-Teacher On-Policy Distillation (MOPD2): After mixed RL, MOPD2 combines autonomous student rollouts with prefix-conditioned…
I work on making large language models practical on constrained hardware through mixed quantization, inference optimization, and serving experiments. Contributions help cover calibration, GPU compute, storage, and testing so these results can be published openly. This is a mixed-precision conversion of XiaomiMiMo/MiMo-V2.6-Flash-RL, pinned to revision 3b38d063180c3e4aed9691fdc735f3d10b266ee4. The source stores routed expert weights as packed MXFP4. Two logical weights occupy each packed byte, so counting stored tensor elements can produce a roughly 159B figure. The expanded language trunk contains approximately 308.78B logical parameters; the root checkpoint including embedded MTP and media…
Scaling Reinforcement Learning Toward Self-Improvement MiMo-V2.6-Pro-RL is the flagship checkpoint of the MiMo-V2.6 series. The series is built to scale reinforcement learning toward self-improvement — scaling RL compute, environment diversity, and grader compute together, so the model keeps expanding its capability frontier through exploration and feedback. Key features include: - Native Omnimodal + Long Horizon: Text, image, video, and audio in one model; 1M tokens for long repositories, tool traces, and multi-session agent runs. - Multi-Prefix Multi-Teacher On-Policy Distillation (MOPD2): After mixed RL, MOPD2 combines autonomous student rollouts with prefix-conditioned single-turn…
Scaling Reinforcement Learning Toward Self-Improvement MiMo-V2.6-Pro-RL is the flagship checkpoint of the MiMo-V2.6 series. The series is built to scale reinforcement learning toward self-improvement — scaling RL compute, environment diversity, and grader compute together, so the model keeps expanding its capability frontier through exploration and feedback. Key features include: - Native Omnimodal + Long Horizon: Text, image, video, and audio in one model; 1M tokens for long repositories, tool traces, and multi-session agent runs. - Multi-Prefix Multi-Teacher On-Policy Distillation (MOPD2): After mixed RL, MOPD2 combines autonomous student rollouts with prefix-conditioned single-turn…
Complete uploaded 5% hot checkpoint: 1325 of 26496 routed experts use NVFP4. Remaining experts use per-expert ARVQ FP4 books and FP16 block scales. Global allocation ranks routing-weighted output-error benefit measured on 65536 training-only calibration tokens. All 69 layers have prepared weights. Sequential PV is continuing through the remaining layers. Layers 1-20 reuse the earlier accepted all-cold candidates (layer 9 retained its initial fit). Their cold weight tensor payloads were verified identical to those underlying the cached inputs for layer 21. Layers 21 and 22 have completed PV; layers 23-69 follow sequentially, freezing the selected NVFP4 hot experts. Each accepted layer…
Model Collections
Hand-picked starting points, each with the reason it exists.
Collection · 4 entries
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.
Collection · 4 entries
Models that fit 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.
Collection · 6 entries
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.
Collection · 3 entries
Speech and audio models
Recognition and synthesis models, grouped so the two directions are easy to compare.
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
The same open-weight model priced by every host that serves it, per million tokens.
Research Hub
Data center trackers and maps
Moratoriums, permits, power, water and capital behind the facilities that run these models.
Method
How the Model Hub is built
Sources, evidence labels, refresh behaviour, and the limits of every comparison here.