Gnihton
4.7K posts
If you're on a Mac, make sure you run `qwen3.6:35b-a3b-nvfp4` or `qwen3.6:35b-a3b-mxfp8`. Those are a lot faster because they'll run on the MLX backend.
ollama@ollama
Qwen 3.6 is here, and open-source! Run it locally with improved agentic coding capabilities. Try it with Claude Code: ollama launch claude --model qwen3.6 Try it with OpenClaw: ollama launch openclaw --model qwen3.6 Run it: ollama run qwen3.6
English
@BountySecurity I'm not very experienced yet, but I'm following your posts. Have these scan profiles, scan scenarios, and vulnerabilities been improved?
English
We surveyed Burp Bounty Pro users. About half didn't know Tags Manager existed.
What tags do:
🔍 Filter 254 profiles by category
👁 Launch active and passive scans by tag
🧠 Power Smart Scan IF-THEN rules
🎯 Organize custom profiles by tech stack
English
开源macOS清理工具PureMac v2.0.0发布
github.com/momenbasel/Pur…
PureMac 2.0近日发布，作为一款专为 macOS 打造的免费系统清理工具（开发者强调，该工具“100% 免费”，不包含试用版、付费高级版或订阅等级 ）。
PureMac 使用 SwiftUI 原生开发，无 Electron 或网页壳层，力求保持界面和性能的原生体验。其采用 MIT 许可证开源，鼓励用户自行审计代码和二次开发。，它以“完全离线运行”和“开源代码可审计”为核心卖点，主打透明和隐私优先的系统维护理念。该项目开源，任何人都可以查看代码，以此强化对用户隐私保护的信任基础，避免常见清理软件在后台收集数据或上传信息的隐忧。
中文
Anyone tried Strix ??
Open source AI hackers to find and fix your app’s vulnerabilities.
Resource: github.com/usestrix/strix (I haven’t tried yet)
English
@lucasmeijer The Google team works like civil servants, and their products are bad. Because they are civil servants, the team gets paid no matter what.
English
I really want these guys to be great, but I'm 30 clicks after seeing this tweet, and I still have not found the actual thing they shipped! I have production apps that could use this, and I give up because it's always such a shitshow to use anything google ai.
Logan Kilpatrick@OfficialLoganK
Introducing Gemini 3.1 Flash TTS 🗣️, our latest text to speech model with scene direction, speaker level specificity, audio tags, more natural + expressive voices, and support for 70 different languages. Available via our new audio playground in AI Studio and in the Gemini API!
English
nvidia'nın nim ai platformuna anthropic uyumlu proxy hazırladım.
claude code üzerinde ücretsiz olarak minimax 2.7, glm-5, kimi k2 gibi modellerle takılıyorum.
nvidia.srv.hermestech.uk
Türkçe
@Daniel_Rubino I assume you'll only be using the browser and X on your computer. Besides, those don't usually consume much battery power.
English
I finally have this review coming soon, but Dell's 2026 XPS 14 is quite possibly the best laptop I've ever used.
Love the keyboard, touchpad, build quality, design, and this is the tandem OLED model with Intel Core Ultra X7 358H, 23GB, and 1TB SSD, which is 🔥
The only thing I could think would be better is the 16-inch, which I'm actually glad omits NVIDIA this time.
Battery life for that config is waaay better than I would have thought (I also have the full HD IPS one, too).
Anyway, if you have any questions for me for the review, thread 'em here.
English
Hepsiburada satıcılı Adidas Ayakkabı ve spor ürünlerinde %50 indirim mevcut.
Daha önce paylaştığımda çok kişi faydalanmıştı ve ürünlerin orjinal olduğunu teyit etmiştik.
Zaten Hepsiburada’nın çakma ürün satma ihtimali yok :)
İndirimli ürünler bağlantısı;
app.hb.biz/pQftBsEMiy0m
#işbirliği
Türkçe
@chams_builds @pcshipp Antigravity IDE is really giving errors. It doesn't matter if you use the Gemini or Claude model. Even those using the Ultra model are experiencing this problem. The same error persists even if you add skills. I think even the team that developed Antigravity doesn't use it.
English
@hadiyapma @pcshipp haha I wish i got money from google.
anyhow, you're saying Claude is giving you limit errors too? what's the Claude plan you're on, and did you try setting up CLAUDE.md, CONTEXT.md , skills etc to save tokens?
English
@chams_builds @pcshipp You must have received money from Google to adopt this idea. I have a Pro account, but when I type a short prompt, I get a limit error, and it immediately gives an error in long conversations. I can't even use Claude at all; it immediately gives a limit error.
English
@pcshipp interesting. I really like Antigravity itself, and think it’s as good as Cursor.
Reason I personally don’t use it regularly is because I already pay for the Claude ecosystem overall (Cowork, chat, Code etc)
What’s the issue you experience with Antigravity?
English
@1slimewell @0xSero Droid is truly the best CLI, but the tokens run out quickly, which is why it's expensive.
English
English
@code_tushar @GoogleDeepMind Probably temporary outage. They are having capacity issues.
English
I downgraded my Google AI Pro subscription, and the downgraded plan was active *immediately*, even though I still had almost one month of Pro plan left.
Needless to say, this left a sour taste in my mouth. Not cool, @GoogleDeepMind 😠
English
@forgebitz You're absolutely right. Even the team that developed Gemini probably doesn't use it. I don't understand how it gets such a high score. It's really bad at coding.
English
@KSolmaz @h_aticedev onların verdiği api key claude opus 4.6 kullandığınıza emin misiniz? bence open llm modellerini kendi sunucularında barındırıp oradan veriyorlar. Modellerin adını değiştitirler. yine de kendi orjinal claude ha da codex gibi verimli/kaliteli sonuç vermez.
Türkçe
@hadiyapma @h_aticedev Ben kullaniyorum iki hesabim var gayet memnunum isine gelmeyen 200 dolar versin ayrica bunu simdi eve baglanip denedim gayet de iyi calisiyor
Türkçe
@KSolmaz @h_aticedev böyle dandik projeler önermeyin. sitesine baktınız mı? Bazı llm modeleri 13-14 saatir down olmuş. bunun yerine poe.com kullanın
Türkçe
latest version 0.1.4 is working perfectly! thanks for the massive support of the community, OpenClaude has now 1.1K stars in Github #OpenClaude
GitLawb@gitlawb
We forked the leaked Claude Code source and made it work with ANY LLM: GPT, DeepSeek, Gemini, Llama, MiniMax. Open source. The name is OpenCode
English
GÖKYILDIZ@GokyildizOnur
Biz de aynı paper üzerinden (TurboQuant, arXiv:2504.19874), aynı günlerde başladık. tq-kv sıfırdan Pure Rust standalone library + inference engine. Çoğu implementasyon llama.cpp fork'u olarak yazılıyor, sadece llama.cpp kullananlar faydalanabiliyor. Biz bağımsız kütüphane yaptık. cargo add tq-kv ile projene eklersin. Rust, C, Python fark etmez. GGUF quantized modellerde herkes patladı biliyorsunuz. Modeli sıkıştırıp küçültünce KV cache compression da üstüne binince çıktı çöp oluyor. 3-Fix framework'ü tam bunun için yazdık. Model sıkıştırılmış halde bile anlamlı cevap veriyor, +17% PPL ile coherent output alıyoruz. Hız: projection 115x, attention 8.9x. 102 test, CUDA + CPU. crates.io'da v0.5.0 yayında. github.com/onur-gokyildiz…
QME
The Opus distilled Qwen 27B is going viral again, but everyone is promoting the old version! You MUST get the v2 version, which Jackrong currently does not have listed in his Opus Fintune collection, so no one is finding it. I discussed the improvements with v2 in my last week's test. Please go check out that post with the video, pinned on my page, if you get a second, and download the v2 here: huggingface.co/Jackrong/Qwen3…
English
CloudFront WAF sets a 403 interception rule for the `/actuator` path, but you can use URL encoding `/%61%63%74%75%61%74%6f%72` (That's, each character of `/actuator` is hexadecimal encoded) to bypass the WAF and directly access Spring Boot
#BugHunter #BugBounty #BugBountyTips #CloudFront
English