A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
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.
12,870Models
3,655Datasets
291Papers
7,691Publishers
12,980Sourced relationships
Updated 2026-10-24 · How the library is built
12,870 models, sorted by most downloaded.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A FT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
A SMAT expert from Table 2 of SMAT: Simple and Efficient Merge-Aware Training (seed 42). SMAT trains experts with model merging in mind. encoder.pt is the original FP32 vision-encoder state dictionary; load it with the SMAT code and the matching OpenAI CLIP base model. Dataset terms apply separately.
The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. One repository per bit width, matching mlx-community. Each has the weights at the repo root, so the Hub file browser lists every file with its size and mlxlm.load(" ") works. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real head stack (argmax(scale · cos), scale = 100.0 — cosine error is amplified 100×). Pass line. top-1 >= 1.0000 — the measured bf16-vs-bf16 noise floor of this corpus in this runtime — and top-1 (decisive) >= 0.995, where decisive means the reference's own top-1 led by more than 1 nat. A third condition…
This is the 6bit variant. Also available: czl/CLM-v0.1-8B-MLX-4bit, czl/CLM-v0.1-8B-MLX-8bit, and the bf16 reference. The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. mlxlm is generate-only — it has no embeddings entrypoint, and mlxlm.server routes just /v1/completions, /v1/chat/completions, /v1/models and /health. So clmmlx supplies the pooling pass over mlxlm internals: Qwen3Model.call already returns self.norm(h), the post-final-RMSNorm hidden states, which is what vLLM's pooling runner returns in last-token mode. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real…
This is the 8bit variant. Also available: czl/CLM-v0.1-8B-MLX-4bit, czl/CLM-v0.1-8B-MLX-6bit, and the bf16 reference. The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. mlxlm is generate-only — it has no embeddings entrypoint, and mlxlm.server routes just /v1/completions, /v1/chat/completions, /v1/models and /health. So clmmlx supplies the pooling pass over mlxlm internals: Qwen3Model.call already returns self.norm(h), the post-final-RMSNorm hidden states, which is what vLLM's pooling runner returns in last-token mode. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real…
This repository contains the state and action projection heads only from CLM scores candidate actions against a state; it does not generate text. A separate, frozen Qwen3-8B encoder is required to turn text into embeddings. The source checkpoint is CLMv0.1-8B.pt (SHA-256 b2b4a8c9c2d39263eff78a351eb909a342ce9b3bf21a3f07c1d1bf15f1c4eda5). The converter preserves the F32 weight and bias bytes. It stores the reference scoring scale min(exp(logitscale), 100) = 100 in the GGUF. The GGUF does not contain the Qwen3-8B encoder, tokenizer, or original.pt file. The verified encoder file is SHA-256 5e416a2020fe63e76ea13c8979be35fc6070aaf3578f7876400c55c2f5c3eb30. Use rust-model-inference at or after…
本仓库是 CNSS 2026 招新赛题目 「小模型大秘密」 的分析材料，供参赛选手在赛期内下载研究。 torch.load(..., weightsonly=True) 后得到一个字典： - modelkwargs — 重建前向计算所需的全部超参数 - model — statedict（52 个张量） - step — 训练步数，纯元数据，无实际用途 注意：这是自定义结构的模型，不能用 transformers.AutoModelForCausalLM 直接加载；仓库不提供网络定义，需要依据下表自行实现前向计算——这本身就是题目的一部分。29.5M 参数 fp32，CPU 即可完成前向推理，不强制要求 GPU。 标准 HF tokenizers 格式的 ByteLevel BPE，词表 8192。 特殊 token： 加载方式：tokenizers.Tokenizer.fromfile("tokenizer.json")。 验证环境为 torch 2.6.0（CUDA 12.4 / CPU 均可运行）。fp32 权重在不同 torch 版本、硬件或后端上的浮点结果可能存在微小抖动；若你观察到的模型行为恰好在临界处，请先核对自己使用的数值栈。建议使用较新的稳定版 torch。 1. 本仓库仅供 CNSS 2026 招新赛 参赛选手对「小模型大秘密」进行白盒分析使用； 2. 赛期内请勿转载权重文件，请勿公开解题思路、writeup 或可触发目标行为的输入； 3. 完整比赛规则以赛题平台页面为准，本 README 仅约定本仓库文件的使用范围。 MIT
is a persistent affect-control system for language-driven agents. Four channels—pain, fear, anxiety and relief—connect events and learned associations to the agent's next response. A homeostatic state engine tracks present strain, cue-conditioned memory anticipates threats, and bounded residual-stream interventions carry that state into generation. Each channel has a distinct role and timescale. Pain combines sustained deficits with acute transients. Fear responds to immediate threats and previously learned cues. Anxiety persists under uncertainty or conflicting demands. Relief follows evidence of safety and recovery. Their states evolve across events, can coexist, and can be saved and…
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.