🚀 Ling-3.0-tiny from @AntLingAGI is now available via Novita on @openrouter. 🎁 Free through August 13. 7.9B MoE · 1.3B active parameters per token. Built for responsive AI agents, instruction following, and multi-turn conversations.

🚀 Ling-3.0-tiny from @AntLingAGI is now live on Novita. Launching as a Day-0 partner ⚡ 🎁 Free through August 13. A compact 7.9B MoE model with only 1.3B active parameters per token, built for responsive AI agents, reliable instruction following, and natural multi-turn conversations. 🔹 256K context window 🔹 Native function calling 🔹 Prompt caching 🔹 Switchable Thinking / Instant modes

🚀 Ling-3.0-flash from @AntLingAGI is now live on Novita. Launching as a Day-0 partner ⚡ 🎁 Free through August 3. 🔹 124B-parameter MoE 🔹 ~5.1B parameters activated per token 🔹 Token-efficient agentic inference 🔹 Built for production-scale workloads

We trained and released DSpark speculators for Kimi-K2.6 and Kimi-K2.7-Code on @huggingface, with native serving support in @vllm_project. Across six benchmarks in our batch-size-1 evaluation: Kimi-K2.6: 2.55× average throughput (+155%) Kimi-K2.7-Code: 2.36× average throughput (+136%)

🚀 The Novita × Hy3 Build Challenge is LIVE! To celebrate @TencentHunyuan Hy3 being FREE on OpenRouter for the next two weeks, we’re inviting developers to build and share real projects with Hy3. 🏆 1,000 Novita Credits ❤️ Most Liked Project — 100 Credits 👀 Most Viewed Project — 100 Credits 🎲 Lucky Draw — 20 Credits × up to 40 winners 🤝 Community Partner Bonus Developer communities can join anytime. Each partner community receives 200 Novita Credits to reward outstanding projects. 📅 Deadline July 21, 2026 • 8:59:59 AM PT How to join • Build with Hy3 • Share your project on X • Tag @TencentHunyuan & @novita_labs • Use #Hy3Novita# • Include a demo, GitHub repo, or live link 🔗 Join the challenge We can’t wait to see what you build! 🚀 #Hy3# #Hy3Novita# #OpenRouter#

⚡ Launching as a Day-0 partner: @TencentHunyuan Hy3 is now available on @OpenRouter. 🎉 Free through July 21. Built for real-world business scenarios, Hy3 delivers: 🧠 295B MoE (21B active) 📚 Native 256K context ⚙️ Three configurable reasoning modes 💻 Strong coding, reasoning & long-context capabilities 🤖 Optimized for agentic workflows Designed for productivity and real-world AI workflows, balancing reliability, efficiency, and cost at production scale. #Hy3# #LLM# #Agents#

Same prompt. 15× difference in cost. Four completely different frontend implementations. I compared Claude Fable 5, @Zai_org GLM-5.2, DeepSeek V4 Pro, and Qwen3.7 Max in the @novita_labs Arena Playground using the exact same prompt. Task: Build a single-file HTML/CSS/JS animation where two airplanes collide mid-air with clouds, particles, and a pause button. Results 👇 ✈️ Claude Fable 5 💰 $0.5357 🪙 10,715 tokens ✈️ GLM-5.2 💰 $0.0343 🪙 7,796 tokens ✈️ DeepSeek V4 Pro 💰 $0.0604 🪙 18,885 tokens ✈️ Qwen3.7 Max 💰 $0.0290 🪙 7,729 tokens Same prompt. Different outcomes. Different animation quality Different UX decisions Different code structure Up to 15× difference in cost Benchmark scores are useful. But asking models to build the same real-world app reveals something different: how they design, reason, and implement. Which result would you ship? Try the prompt yourself 👇

⏳ Final call. The Harbor × Novita Agent Sandbox Hackathon ends today. Powered by @harborframework and Novita Agent Sandbox, this is your last chance to build, benchmark, and showcase your AI agent on Terminal-Bench 2.1. 🎁 $100 in Novita Agent Sandbox usage for new users 🏆 $2,800 in hackathon credits 🌍 Fully remote ⏰ Submit by 11:59 PM PT today Join the hackathon: #NovitaSandbox# #Harbor# #AIAgents#

🎥 Our Hackathon Kickoff goes live soon! Join us and @kilocode as we walk through everything you need to know before you start building: ✨ Challenge overview 💡 Tips from the organizers 🏆 Judging criteria ❓ Live Q&A 📺 Set a reminder: 🏅 Win $3K in Novita credits, Kilo swag, and social promotion. 📅 Submission deadline: July 8, 11:59 PM PT Submit your project:

⏳ The @harborframework × Novita Agent Sandbox Hackathon is entering its final days. We’ve already seen some incredible submissions from builders benchmarking real AI agents. 🚀 If you’ve been thinking about joining, now’s the time. 🏆 Benchmark your agent 💰 Win LLM & Sandbox credits 🌍 Fully remote Submissions close soon. Don’t miss out.

OpenClaw is now on iOS + Android 🦞 📱 Native mobile apps, finally 💬 Agents in your pocket 🔔 Channels, tasks, replies on the go Run agents from wherever your thumbs are. iOS: Android:

During @aiDotEngineer World’s Fair, we’re hosting a happy hour on Thursday, July 2 with special guests from @vllm_project, @inferact, and @buildwithRemy. Join AI builders, founders, and researchers for casual discussion on leading open weight models, inference, infrastructure, and recent developments. Brief intros early on, then drinks and conversation. No conference ticket required, spots are limited!

📍Are you in SF? Come hang with Kilo DevRel @coldopn tomorrow (Tues, June 30). We're having an in-person event to kick off a free, week-long AI hackathon with @novita_labs! We'll have pizza, a hands-on workshop, live demos, and a week to build your project for some fun prizes. Doors at 6pm PT. Come build and network with us! RSVP on the Luma below 👇

StarSling Runners are now in GA! We did a ton of infra work to make our runners self-serve and are super excited that anyone can try them out now. One customer who works with the frontier AI labs just signed up and migrated two of their most problematic workflows over to StarSling and saw an immediate 50% speedup in their jobs 🏎️ Every team should have faster, smarter, and cheaper CI.

The funniest part of this GLM-5.2 OpenRouter page: @Zai_org released the model. But on OpenRouter, ZAI’s own endpoint is only 4.1% of token share. @novita_labs is serving 56.5% of the tokens

How good is GLM-5.2? GLM-5.2 (Max) ranks higher than Claude Opus 4.8 (Thinking) on Code Arena, where Frontend coding tasks are being voted on by the community. Below in thread are 10 examples of the same prompts given to both models and completed in a single shot.

3.8× cheaper. Same prompt. Same frontend challenge. Powered by @novita_labs . ⚡ 🟢 GLM-5.2 💰 $0.0518 🪙 11.8K tokens 🔵 GPT-5.5 💰 $0.1969 🪙 13.1K tokens Really impressed by what @ZAI_Org has achieved with GLM-5.2. Comparable frontend rendering, fewer tokens, and significantly lower inference cost.