Private fleet archive. Canonical recipe: reports/2026-09-23fleetcloudarchive/legacy/salvage-h20-code-20260921 Use the exact recorded revision and verify SHA256SUMS. This package is a snapshot, not a live directory mirror.
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.
11,734Models
3,323Datasets
288Papers
7,150Publishers
12,282Sourced relationships
Updated 2026-10-20 · How the library is built
11,734 models, sorted by most downloaded.
Built with Qwen. Research and evaluation only, non-commercial. This is the rank64 EMA adapter from qwen21-m11-r64/ckpt-3000-ema. It redraws a furniture carcass as an orthographic front elevation: solid black boards and white openings. It is an adapter, not a complete model. It does not infer real width, depth, thickness, hardware, hidden joints or safe construction. GPT Image 2.5: frontal edit → board drawing. m11: photo → board drawing. Both actual drawings are compiled by the same Fabivo reader with no repaired boards. Three saved examples compare these different pipelines, not matched inference budgets or independent unseen accuracy. GPT uses low quality and frontal candidate 1; m11 uses…
Domain TopK sparse autoencoders trained on the residual stream of instruction-tuned LLMs while they read fact-checking verdict prompts (claim + evidence + instructions), released to study how models decide to abstain ("NOT ENOUGH EVIDENCE") instead of issuing a verdict. Off-the-shelf SAEs reconstruct the verdict-decision position poorly (it is out of their training distribution); these SAEs are trained on the task distribution itself, so the decision position is in-distribution. Training follows Gao et al. (2024) — see each SAE's card for the full configuration and metrics. Browse the latents in the SAE viewer: lucasfrag/fact-checking-abstention-saes-viewer (top-activating claims at the…
YOLO weights for factory-safety-isaac — a body-camera-only factory safety agent built on NVIDIA Isaac Sim. These are the two trained detection models the project uses; no other model files are hosted here (MediaPipe's hand landmarker and the Qwen2.5-7B LLM are pulled from their own upstream sources at first run, see the project README). on 7,700 synthetic images generated in Isaac Sim (Replicator auto-labeling), 70% bodycam / 30% free-viewpoint. Or with the project's own scripts (scripts/runpatrol.py --weights, scripts/runpatrol.py --pinch-weights ). Trained entirely on synthetic data (NVIDIA Isaac Sim warehouse assets + Poly Haven CC0 tool scans). See the main repository for full…
An archive of a market-making project that did not establish a learned, independently verified live trading edge. It includes the weights, the promising results, the losses, and the evidence that changed the interpretation of those results. The project explored Polymarket BTC UP/DOWN five-minute markets using Polymarket order books, Binance market data and Chainlink observations. It evolved from a replay engine and feature pipeline, through supervised signals and offline policies, into an Alpha + MakerPolicy deployment and a newer architecture. This release preserves that progression without rebuilding the training or trading system. There are 24 retained trained candidates: 3 M3, 1 M5…
Koleksi checkpoint model dan LoRA adapters untuk penalaran hukum pidana Indonesia berbasis putusan peradilan tingkat pertama Mahkamah Agung Republik Indonesia. Proyek riset ini mengevaluasi adaptasi domain peradilan (Supervised Fine-Tuning & QLoRA) dari model berskala ringan (1.5B/3B) hingga skala besar (12B, 27B, dan 31B) untuk menyusun pertimbangan hukum (ratio decidendi) dan amar putusan (dictum) yang setia pada fakta persidangan dan patuh pada asas pembuktian Pasal 184 KUHAP. Repository ini diorganisasikan sebagai satu project induk bertingkat (multi-folder model repository), di mana setiap arsitektur model, checkpoint per-epoch, dan artefak evaluasi dikelompokkan ke dalam subdirektori…
Checkpoint collection for four real-world tasks and four model families. - joint is a top-level folder alongside the four tasks, for models trained on all tasks. - Task-specific runs sit directly under / /. Each run includes its original checkpoint format and inference/training metadata. Reserved folders contain a README; inspect run folders for available weights. See CHECKPOINTUPLOADS.md for the upload contract and commands, checkpointlayout.json for the path schema, and hfcheckpointimports.json for pinned source revisions and file hashes. Imported run READMEs retain their original training and evaluation limitations; these historical runs are not asserted to use the current benchmark…
Action-conditioned robot world model (FastWAM MoT: Wan2.1-T2V-1.3B video branch + action DiT), trained on DreamZero-DROID. This is the PhyPO run droidphypo1p3b (runs/droidwan211p3b1e-4/2026-09-0801-06-31) at step 100,000. The video DiT was initialized from the PhyPO/GRPO-finetuned Wan2.1-T2V-1.3B (Wan-AI-PhyPO-1.3B), and the action expert from ActionDiTphypoWan211p3Balphascale1024hdim.pt. This is not the stock Wan2.1 baseline (droid1p3bscratch). - step100000.pt — model weights (3.9 GB). Dict with keys mot, proprioencoder, step, torchdtype. - datasetstats.json — action/proprio normalization stats; required for inference. - The same checkpoint is also stored at…
Intermediate inference checkpoint from the v10 two-stage run started 2026-10-15. This release contains joint step 2,000, following 1,000 steps of new-module warmup. It is not the warmup checkpoint, and no online RoboMME success rate is claimed. joint/step002000/ contains config.yaml, datasetstats.json, provenance.json, SHA256SUMS and checkpoints/weights/step002000.pt. The single.pt contains the trained MoT (including v10 modules), proprio encoder and model metadata. A separate v7/warmup adapter checkpoint is not needed. Frozen external assets such as the VAE and text encoder are not bundled; use the compatible FastWAM installation and adjust machine-specific config paths. Optimizer…
Reading notes and an experiment sketch for Few Shot Multimodal. The repository emphasizes what still needs to be tested instead of manufacturing scores or release claims. - the scope of the research question and likely confounders - a proposed comparison with matched baselines - concrete evaluation context such as task-appropriate public benchmarks named in the main note - reproducibility checks, failure modes, and open questions - topic-relevant references Start with papernotes.md for the full note. Sections labeled as plans or hypotheses should not be interpreted as experimental results. If results are added later, they should include dataset versions, commands, seeds, hardware, and raw…
Reading notes and an experiment sketch for Few Shot Multimodal. The repository emphasizes what still needs to be tested instead of manufacturing scores or release claims. - the scope of the research question and likely confounders - a proposed comparison with matched baselines - concrete evaluation context such as task-appropriate public benchmarks named in the main note - reproducibility checks, failure modes, and open questions - topic-relevant references Start with review.md for the full note. Sections labeled as plans or hypotheses should not be interpreted as experimental results. If results are added later, they should include dataset versions, commands, seeds, hardware, and raw logs.…
This is an exploratory note for Few Shot Multimodal. It records the intended comparison, likely confounders, and reproducibility requirements before any benchmark result is reported. - the scope of the research question and likely confounders - a proposed comparison with matched baselines - concrete evaluation context such as task-appropriate public benchmarks named in the main note - reproducibility checks, failure modes, and open questions - topic-relevant references Start with reading.md for the full note. Sections labeled as plans or hypotheses should not be interpreted as experimental results. If results are added later, they should include dataset versions, commands, seeds, hardware…
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.