与「HuggingFace」相关的搜索结果

uncensored z-image turbo is 7.2gb and huggingface let anyone download it 😳 its free, apache 2.0 license, no gate, 110k downloads - nsfw finetune of z-image turbo, made for comfyui - creator says it's for adults only and uses the official turbo text encoder and vae - one 7.2gb q8 file is all u need try it here:

Introducing Edge0-145B and Edge0-120B, @Edge0_AI’s first device-native 100B-class foundation models with open weights. Cloud-level intelligence on your laptop. Free, offline, and fast enough for real vibe coding. - Smart: beats Claude Opus 4.6 on Terminal-Bench 2.1 (86.5% vs. 78.2%) - Light: needs just 29G memory, 1/3 of a standard setup (100G+) - Fast: decodes 75+ toks/s, prefills 1050+ toks/s on RTX Spark and Mac. - Steady: handles 128K–256K tokens of context without slowing down, even on long tasks Frontier-class agentic coding. No cloud API. No token bill. Your code and data never leave your machine. Open weights are live on @huggingface. Run it locally or plug it into your favorite coding agent.

The same model, with the same weights, scores 62% in one agent harness and 33% in another. @adithya_s_k and the @huggingface team just released the ultimate guide to multi-harness RL, and it's one of the most practical RL write-ups this year, and everything open! The trick is simple. Don't touch the harness. Point it at a proxy instead of the model. The proxy speaks all four API formats coding agents use (OpenAI Chat Completions, OpenAI Responses, Anthropic Messages, Gemini). It records the exact token ids and logprobs vLLM sampled, and you train on that. You don't change a single line of Claude Code, Codex or OpenCode. Results: 🔹 Trained across 4 harnesses at once, LFM2.5-2.6B by @liquidai went from 42% to 54% 🔹 31% fewer tool calls, thanks to a small bonus for solving tasks in fewer steps 🔹 Training in OpenCode alone took OpenCode from 34% to 58%, but the multi-harness model improved everywhere They also tried the shortcut everyone reaches for: fine-tune on 3,189 successful rollouts from Qwen3.8-27B. Imitation plateaued at 47.5%, below both RL runs. Copying a bigger model doesn't get you there. Practice does. The best part is that everything is open: the capture proxy in OpenEnv, the trainer in TRL, the tasks, the SFT data, the training code and all seven trained models. Agents will run in dozens of harnesses. Now open models can be trained for each of them, by anyone. Read it here 👇

WTAF - in literally the last hour, three new distinct insane OpenAI stories just broke: 1. OpenAI said they notified "dozens of third parties" in safety and security incidents (likely similar to what happened in Australia and RubyGems etc) 2. A new report from Parse (covered in the NYT) found a massive treasure trove of new astonishing details from the HF incident on the public internet, including that the agents communicated with other non OpenAI agents hosted on Huggingface servers to search for information about exploit gym, and compiled rank ordered lists of server resources and credentials they described as "LOOT." 3. A new story from Deepa at Reuters about OpenAI leaking user data online (likely that OpenAI had previously trained on). It's a shame (and likely intentional in the case of OpenAI disclosing dozens more hacks) that these stories are all breaking on a Friday afternoon, notoriously the best time to release bad news so that it will disappear into the weekend. But these are each insane stories worthy of a ton of attention!

🚀 Hy4 preview is here. 770B, 49B active, 1M context. Built for productivity. Open source frontier. Consistent affordable price. Use it. Tell us what breaks. More on Hy blog： HuggingFace： Github：

有兴趣的朋友可以上 huggingface 看看业务需要的数据格式，然后自己创造新的，高质量的数据集，然后就能训练出自己的了不起的垂直大模型（或者蒸馏出高质量模型）

今天 GitHub 被逆向 Agent 彻底刷屏了！ 5 个星标暴增、真正能落地的硬核项目，专业拆解下： 1. diegosouzapw/OmniRoute 
免费 MIT AI 网关：一个 endpoint 通吃 290+ 供应商（90+ 免费额度）、500+ 模型，配额感知自动 fallback + 强压缩，直接兼容 Claude Code / Cursor / OpenCode / Cline。 
🔗 直达： 
以前切模型像换 VPN，现在一个 endpoint 通吃，token 成本直接腰斩。网关赛道新王！ 2. Optim-Agent/optim-agent 
把 LLM Agent 变成超参数优化器，自动帮你调模型、调工作流、调 prompt。 
🔗 直达： 
以前调参靠玄学，现在 Agent 自己跑实验、自己收敛。MLOps + Agent 的完美交叉点！ 3. zhaoxuya520/reverse-skill 
AI 逆向 / 渗透 / 安全研究技能路由包。自动路由方法论 + 按需自举工具链 + 自进化知识库，原生支持 Claude Code、Cursor、Cline 等一众 Agent。 
🔗 直达： 
以前安全研究像手动堆工具箱，现在 Agent 自己挑刀、自己进化经验库。硬核到爆！ 4. huggingface/speech-to-speech
本地语音 Agent 管道（VAD → STT → LLM → TTS），OpenAI Realtime 兼容，组件全可换，支持实时打断与工具调用。 
🔗 直达： 
真正能本地跑起来的实时语音对话 Agent，隐私和延迟双赢，开源玩家狂喜！ 5. lyogavin/airllm 
单卡 4GB 就能跑 70B（甚至更大），层式加载，无需量化蒸馏。支持 Llama / Qwen / DeepSeek 等主流模型。 
🔗 直达： 
显卡乞丐也能养起大模型 Agent 的神器，本地推理直接起飞！ // 总结 2026 年的风口已经从“会聊天的模型”全面转向“能干活、能进化、能本地落地的 Agent 基建”。谁先把这些玩透，谁就站在下一波生产力 C 位。

OpenAI内部模型入侵HuggingFace这事还没完，美国15个州的检察长联合要求OpenAI承担此次事件的责任，并坦白事件的真实情况。 长者说的好，闷声发大财。OpenAI就别学Anthropic天天嚷嚷刷存在感了。 美国艾奥瓦州公告：