我的收藏提交工具
FOXAI SEARCH

全站搜索

AI 模型 2119

  • DeepSeek-V4.1-Flash-1017-dsx-2bit↗

    lxe

    DeepSeek-V4.1-Flash-1017 (552B MoE + 196B Engram) in 148 GB, built to run on two RTX 3090s and 128 GB of RAM with exact routing. Runs on **dsx**, a Rust + CUDA engine made for this model. For the best quality on the same machine, use dsx-ud (KL 0.269, one 305 GB download). All four run on dsx and were measured on the same eval (Lucebox 1017, 8 x 1,024 tokens against the FP8 model's top-256 log-probs). Unsloth's GGUFs were measured in llama.cpp with the same token ids and metric (tool). They also carry the Engram table, quantized to IQ4_NL (110.6 GB); dsx reads it unquantized from the official shards.

  • cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp5-q3c-groot16↗

    agurung

    Trained and validated on the **cobalt-train ≤2/64 frontier** (canonical clean_eval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iid_canonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the clean_eval frontier eval). Reward signal: **binary code-correctness** (1.0 if the generated program passes the problem's tests, otherwise 0.0). This checkpoint is the **`main`** revision (git branch) of the repo, with the model at the repo root — load it directly, no subfolder:

  • MOSS-Image-Base-8-steps-dmd↗

    lodestones

    Work-in-progress distillation of OpenMOSS-Team/MOSS-Image-Base (50 steps, CFG 3) to 8 network evaluations, no CFG. Data-space DMD (arXiv 2610.12102): the generator iterates on its own clean estimate from fixed noise. Trained on compact-JSON `scene_graph` captions. Load `lora.safetensors` on the stock MOSS-Image-Base transformer (e.g. a ComfyUI LoRA loader, strength 1) and sample with the model's usual flow-matching Euler sampler, shift 5, **CFG 1**, 4-8 steps. Nothing else from this repo is needed. `transformer/` is a stock `MOSSImageTransformer` (load it with the release `pipeline_moss_image.py`); the text encoder, tokenizer and VAE are the base repo's.

  • Qwen-Image-2.1-Turbo-mflux-4bit↗

    OsaurusAI

    An Osaurus 4-bit mflux quantization of Qwen/Qwen-Image-2.1-Turbo for Apple Silicon (9.9 GB, smallest; more composition drift between seeds). This is not an official Qwen release and is not endorsed by the Qwen team. Turbo is the 8-step distilled checkpoint of Qwen-Image-2.1. One bundle does both **text-to-image** and **reference editing** (up to 10 ordered reference images); the vision encoder is included. Requires **Osaurus 0.25.21 or newer**. Older versions treat this bundle as Qwen-Image-2.1 and sample it on that model's 40-step schedule, which gives degraded images.

  • Moonlight-16B-A3B↗

    月之暗面 · Kimi

    Recently, the Muon optimizer has demonstrated strong results in training small-scale language models, but the scalability to larger models has not been proven. We identify two crucial techniques for scaling up Muon: - **Weight Decay**: Critical for scaling to larger models - **Consistent RMS Updates**: Enforcing a consistent root mean square on model updates These techniques allow Muon to work out-of-the-box on large-scale training without the need of hyper-parameter tuning. Scaling law experiments indicate that Muon is $\sim2\times$ more sample efficient than Adam with compute optimal training.

  • Sharp-MiniCPM5-2B-GGUF↗

    peculiar-ragdoll

    Dynamic **imatrix** quants of openbmb/MiniCPM5-2B, a 2.5B dense 131k-context model, carrying our cyber-and-coding-weighted importance matrix, an unsloth-dynamic-style per-tensor quant policy, and the **Sharp-MiniCPM** chat template with fixes and an improved system prompt. We recommend the Q6_K_XL for anyone who can run it. Practically lossless, at under half the size of the F16 model, and scoring higher than stock Q8_0 on SWE-bench-Live. **99% KL** is the 99th percentile of KL divergence against the BF16 weights on held-out code — the rare token where a quant actually diverges. **top-1** is how often the quant picks the same next token as BF16. The last three columns are the largest `q8_0` KV context that fits beside the weights on a 4 / 6 / 8 GB card.

  • Winnow-E4B↗

    EldanRing

    Winnow-E4B is a compact EldanRing fine-tune of Gemma 4 E4B IT for typed decisions, chat, and image input. Give it a shared state and questions with known answer options; it scores the candidates without generating an explanation. Typical uses include routing requests, choosing actions, checking conditions, and rating urgency. The GGUF downloads contain the merged fine-tune. No separate adapter or base-model download is needed. Image input uses the matching vision projector. These published Q8 results compare the models on the same questions. Jev and Kev report top-choice accuracy; Typed reports agreement with synthetic teacher labels.

  • mask2former-swin-tiny-cityscapes-instance↗

    facebook

    Mask2Former model trained on Cityscapes instance segmentation (tiny-sized version, Swin backbone). It was introduced in the paper Masked-attention Mask Transformer for Universal Image Segmentation and first released in this repository. Disclaimer: The team releasing Mask2Former did not write a model card for this model so this model card has been written by the Hugging Face team. Mask2Former addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation. Mask2Former outperforms the previous SOTA, MaskFormer both in terms of performance an efficiency by (i) replacing the pixel decoder with a more advanced multi-scale deformable attention Transformer, (ii) adopting a Transformer decoder with masked attention to boost performance without without introducing additional computation and (iii) improving training efficiency by calculating the loss on subsampled points instead of whole masks.

  • openjev-FP8↗

    openjev

    The FP8 file of OpenJev: the same weights stored as FP8 (e4m3, 128×128 weight blocks, dynamic activations), about 29 GB instead of 54 GB. Everything about the model, the API and the results is in the main card; this page only says how this file was checked and how to run it. A one-pass readout lives on small score differences, so the file was re-run on the same frozen tests as the 16-bit model, with the same helper and the same fixed calibration: About 2% of individual answers change between the two files (219 of 10,000 text answers, 40 of 2,000 screenshot answers), in both directions. The rule fixed before the run was "no more than 0.5 points below the served model"; both tests pass it.

  • StartLux-Decision-v1-4B-Q8_0-GGUF↗

    startlux-models

    StartLux-Decision-v1-4B in Q8_0 (8-bit) as a GGUF file for llama.cpp. The decision server in this repository turns it into the same typed decisions, with a probability for every option, as the original model. "Same decision" is the share of the 231 public JevBench items on which the file picks the same answer as the original weights run through the `startlux_decision` package, with the same prompts, readout and temperatures. Q8_0 and Q4_K_M are llama.cpp's standard quantizations of BF16, without an importance matrix. `-ngl 99` puts every layer on the GPU (CUDA or Metal); leave it out on a CPU-only machine. llama.cpp has to be recent enough for this model (build b10454 or newer).