全站搜索

AI 模型 2119

• gliner2.5-small-v1↗

  fastino

  > *Extract entities, classify text, parse structured records, score span attributes, and extract relations — all in one boundary architecture.* GLiNER2.5 Small is the compact English boundary checkpoint. It keeps the same public API as the larger 2.5 models while running faster on CPU and edge-style deployments. Load it with `AutoExtractor`: the checkpoint's `architecture` field selects `BoundaryExtractor` automatically. Fine-tune via Fastino. Join discussions Reddit.

• Intern-Decision-4B↗

  internlm

  **Intern-Decision-4B** is a multimodal structured decision model fine-tuned from **Qwen3.5-4B**. It accepts a shared state, a schema of named questions, and optional images, and returns an answer distribution for every question in one model forward pass. 1. Preserve the question and option order, and map each question's options to single-token symbols `A`, `B`, …, `Z`, `a`, …, `z`, `0`, …, `9`. 2. Render the original system prompt, state, decision schema, and a complete assistant JSON skeleton with one ` ` placeholder per field. Preserve the checkpoint's chat template and empty thinking block. 3. Run one causal Hugging Face forward pass. For the masked-next-token decision objective, read logits at the position **immediately before each placeholder**. 4. Take a softmax over only that field's allowed candidate-symbol logits, then apply the checkpoint's probability calibration. 5. Map symbols back to the original option values and return typed JSON answers. This API performs structured candidate scoring. It does not call `generate()` or sample free-form text. A request can contain multiple fields; no gold answers are inserted into the prompt. The inference compiler uses only `state`, `questions`, and optional `images`.

• samantha-1.1-westlake-7b-i1-GGUF↗

  mradermacher

• wav2vec2-large-xlsr-53-basque↗

  stefan-it

  Fine-tuned facebook/wav2vec2-large-xlsr-53 in Basque using the Common Voice. When using this model, make sure that your speech input is sampled at 16kHz. Many thanks to the OVH team for providing access to a V-100 instance. Without their help, fine-tuning would not be possible! I would also thank Manuel Romero (mrm8488) for helping with the fine-tuning script!

• wav2vec2-conformer-rel-pos-large-960h-ft↗

  facebook

  Wav2Vec2-Conformer with relative position embeddings, pretrained and **fine-tuned on 960 hours of Librispeech** on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20. This code snippet shows how to evaluate **facebook/wav2vec2-conformer-rel-pos-large-960h-ft** on LibriSpeech's "clean" and "other" test data.

• Speaker-wavLM-pro↗

  Orange

  This model produces embeddings that globally represent the non-timbral traits (prosody, accent, ...) of a speaker's voice. These embeddings can be used the same way as for a classical speaker verification (ASV): in order to compare two voice signals, an embeddings vector must be computed for each of them. Then the cosine similarity between the two embeddings can be used for comparison. The main difference with classical ASV embeddings is that here only the non-timbral traits are compared. The model has been derived from the self-supervised pretrained model WavLM-large. The following code snippet uses the file spk_embeddings.py to build the architecture of the model. Its weights are then downloaded from this repository.

• Ling-3.0-flash-GGUF↗

  AtomicChat

  Ling 3.0 flash is a 124B MoE with 5.1B active parameters; its GGUF chat template includes thinking mode and tool calling. Our AD layouts have 31–41% lower mean KL than stock llama.cpp presets at comparable sizes. The raw measurements and controls are public. From the team behind Atomic Chat , the local AI app for chat and agents. These Ling files require our TurboQuant build; use the CLI instructions below . The tests documented below cover the CLI workflow, not app compatibility. Quantizations of `inclusionAI/Ling-3.0-flash`: 124B total, 5.1B active, hybrid linear attention (35 KDA blocks interleaved 5:1 with 7 gated MLA blocks) over a 512-expert MoE. Bits are placed by hand rather than by the default rules, and the controls that prove it is worth something are published next to the files. At the same size, our layout sits 31 to 41 % closer to BF16 than what `llama-quantize` produces on its own

• LTX-2.5-22b-distilled-nvfp4-comfy↗

  BennyDaBall

  The full NVFP4 build of the LTX-2.5 22B distilled transformer, stamped so ComfyUI actually loads it. Smallest working 22B on the board — **18.7GB, ~2.8GB lighter than the official comfy-int8** — and the same-seed output holds up frame for frame. Top row: text-to-video. Bottom row: **image-to-video** from a single keyframe — same start image, same seed, both quants. Native synced audio on all four (yes, the disco clips come with their own disco track). Lightricks ships an NVFP4 export of the 2.5 distilled DiT, but it's a raw TensorRT-ModelOpt export. Load it through ComfyUI's UNETLoader and you get:

• samsung_moss↗

  Danrisi

• flux-text-encoders↗

  unsloth

  The two text-encoder files FLUX.1 needs to run under stable-diffusion.cpp, mirrored from comfyanonymous/flux_text_encoders. Unsloth Studio fetches these on the no-GPU (native sd.cpp) path. It used to fetch them straight from the community repack; nobody there owes anyone availability, and a rename or a takedown turns every CPU FLUX.1 load into a 404. So they are mirrored here. **The weights are unmodified:** byte for byte the files of the same name in the source repo. Only these two files are mirrored. The source repo also ships fp8 T5 variants; Studio does not use them, so they are not here.