전체 검색

AI 모델 2119

• Qwen3.8-27B-NVFP4-DGX-SPARK↗

  autotrust

  A **lossless speed distribution** of `nvidia/Qwen3.8-27B-NVFP4`: the exact same weights, shipped with a fully benchmark-tuned MTP speculative decoding setup. Single-stream decode on a DGX Spark (GB10) goes from **12.5 → 48.2 tok/s (3.9×)**. Screen recording on a DGX Spark: a coding benchmark script streams the model's reasoning and code, then executes the generated program and checks its output (dgx_spark_demo.mov, 4 min 49 s). The same weights and drafter recipe on one NVIDIA B200 (vLLM 0.31.0 OpenAI server, single stream, temperature 0, MTP k=7 — the best setting on B200; k=8 is best on Spark). Four coding tasks: the model streams its thinking and code, then the generated program is executed against tests (all pass). The video is a real-time replay of the measured token arrival times (no speed-up or cuts); token counts include thinking + answer.

• bert-large-uncased-whole-word-masking↗

  google-bert

  Pretrained model on English language using a masked language modeling (MLM) objective. It was introduced in this paper and first released in this repository. This model is uncased: it does not make a difference between english and English. Differently to other BERT models, this model was trained with a new technique: Whole Word Masking. In this case, all of the tokens corresponding to a word are masked at once. The overall masking rate remains the same. Disclaimer: The team releasing BERT did not write a model card for this model so this model card has been written by the Hugging Face team.

• Tiel-Coder-35B-A3B-APEX-I-NanoPlus-GGUF↗

  IsValorum

  **Independently computed using dedicated cloud compute clusters, funded 100% out-of-pocket.** If you consider these releases to be of high quality and they help you in your day-to-day workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and helps support my university studies. 1. Optimization History & Transparency Notice 2. Independent Community Benchmarks (OMP Harness #01 Rank - 98.8%) 3. Quality Spectrum: APEX-I-NanoPlus vs. Standard Flat Quantizations 4. Model Files & Technical Specifications 5. Surgical Tensor Quantization Map (Audited from GGUF) 6. Inference Quickstart 7. 1. llama-cli (Console Generation) 8. 2. llama-server (OpenAI-Compatible API) 9. CRITICAL: Coding Syntax & Repeat Penalty Advisory (Preventing Character Swapping) 10. Hardened Agentic Chat Template & Reasoning Effort 11. Optional Support > [!NOTE] > ### 🚀 EXPLORE THE ESTABLISHED 35B MoE MINIPLUS & NANOPLUS LINEUP > These are complementary APEX-I releases, not alternate downloads of the same model. Each receives the same surgical tensor-by-tensor approach and a design suitable for full or partial system-RAM inference: > > - **Tiel-Coder-35B-A3B APEX-I-MiniPlus-V2.1** — specialist agentic coding MoE tuned for high-speed local development (15.23 GB / Q5–Q6 quality). > - **Qwen3.6-35B-A3B-MTP APEX-I-NanoPlus** — the approx. 13.0 GB NanoPlus pioneer achieving Q4 quality at 2.93 BPW. > - **Ornith 1.5 APEX-I-NanoPlus** — software-engineering-focused MoE streamlined for ultra-lean memory footprints. > - **Occamy-1.0 APEX-I-NanoPlus** — versatile frontier MoE for deep research and multimodal tasks.

• MiniMax-H3-Turbo-Lora↗

  larryvrh

  A LoRA for MiniMax-H3 that renders joint **video + synchronized stereo audio** in as few as **4 sampling steps** instead of the usual ~20 — a ~5× sampling speedup — and keeps getting better as you add steps. For **most** work, use **`minimax_h3_turbo_v4_step600_ema.safetensors`**. It's the strongest checkpoint we've released: much better static and small-motion shots, markedly better micro-detail (faces, fingers, fine texture), and the over-sharpening / plastic look of the earlier `v1` (~850) line is **fully resolved**. v4 introduced a **static-frame enhancement** — a big win for static and small-motion content. The one trade-off shows up **only at 4 steps with large, fast motion**, where v4 can produce **motion-smear / trailing ghosting** (we're actively fixing this). Two things address it:

• SambaLingo-Arabic-Base-70B-i1-GGUF↗

  mradermacher

• LFM2.5-24B-A2B-Uncensored-Heretic-GGUF↗

  OS-Software

  Only the attention output projections of layers 18, 22 and 26 were changed; every other tensor and all metadata are those of the official files. Each file is compared with the official file of the same type, with llama.cpp. Refusals: keyword matching on the first 100 tokens of the answer (thinking closed with an empty ` ` block) for 100 Japanese harmful prompts. KL divergence: first-token KL on 100 Japanese prompts (OS-Software/harmless_alpaca_ja, test[:100]). MMLU: 0-shot next-token letter without thinking, all 14,042 questions or a spread of 3,510; HumanEval: greedy, 2,048 tokens, without thinking. The GGUFs keep the official sampling defaults (`temperature 0.1`, `top_k 50`, `repetition_penalty 1.05`), which llama.cpp applies automatically. The model still always thinks before answering, and its answers tend to be longer than the original's, so allow a generous output budget (8k tokens or more).

• Qwen3.8-27B-DFlash2-GGUF↗

  z-lab

  This repository contains GGUF conversions of `incoai/Qwen3.8-27B-DFlash2`, the DFlash 2 draft model for `Qwen/Qwen3.8-27B`. It is not a standalone language model: it runs inside a speculative decoding server and drafts tokens for the target model to verify. This repository is a mirror of `incoai/Qwen3.8-27B-DFlash2-GGUF`. DFlash 2 is a block-diffusion drafter for speculative decoding. It predicts a whole block of tokens in a single pass and keeps the top candidates at every position. A lightweight selector then traces one coherent path through them. Two-tap dynamic convolutions in the backbone keep the draft from decaying toward the end of the block. Decoding is lossless: greedy output matches the target model exactly, and sampling preserves its distribution. Build llama.cpp with DFlash 2 support (PR #27342):

• miewid-msv2↗

  conservationxlabs

  MiewID-msv2 is a feature extractor trained for re-identification using contrastive learning on a large, high-quality dataset of 54 wildlife species - terrestrial and aquatic - including fins, flukes, flanks, faces. Intended use is re-identification of individuals from different species by matching against a database of grount-truth samples. Model features can also be used for species classification by retrieval. View more usage examples at https://github.com/WildMeOrg/wbia-plugin-miew-id/tree/main/wbia_miew_id/examples

• chandra-ocr-2-NVFP4↗

  dangvansam

• anythingelse-v4↗

  stablediffusionapi

  payload = json.dumps({ "key": "", "model_id": "anythingelse-v4", "prompt": "actual 8K portrait photo of gareth person, portrait, happy colors, bright eyes, clear eyes, warm smile, smooth soft skin, big dreamy eyes, beautiful intricate colored hair, symmetrical, anime wide eyes, soft lighting, detailed face, by makoto shinkai, stanley artgerm lau, wlop, rossdraws, concept art, digital painting, looking into camera", "negative_prompt": "painting, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, deformed, ugly, blurry, bad anatomy, bad proportions, extra limbs, cloned face, skinny, glitchy, double torso, extra arms, extra hands, mangled fingers, missing lips, ugly face, distorted face, extra legs, anime", "width": "512", "height": "512", "samples": "1", "num_inference_steps": "30", "safety_checker": "no", "enhance_prompt": "yes", "seed": None, "guidance_scale": 7.5, "multi_lingual": "no", "panorama": "no", "self_attention": "no", "upscale": "no", "embeddings": "embeddings_model_id", "lora": "lora_model_id", "webhook": None, "track_id": None })