标签：#AI

Kimi K3 took first place in Arena's Frontend Code evaluation the week it launched, and it holds a 1M-token context — but Moonshot doesn't ship a coding agent, and your coding agent doesn't ship Kimi K3. Claude Code is locked to Anthropic's API by default, Cursor to its own backend, Cline to whatever key you hand it. LLM Gateway bridges that gap. It speaks both the Anthropic and OpenAI API formats, so the tools you already use can run Kimi K3 — or any of 200+ models — with a base-URL change. Here is the exact setup for each tool. Kimi K3 in Claude Code Claude Code talks to any endpoint that speaks Anthropic's /v1/messages format, which LLM Gateway does natively. Three environment variables: export ANTHROPIC_BASE_URL = https://api.llmgateway.io export ANTHROPIC_AUTH_TOKEN = $LLM_GATEWAY_API_KEY export ANTHROPIC_MODEL = kimi-k3 claude That's the whole migration. Every request now routes through LLM Gateway to Kimi K3, and every request shows up in your dashboard with its exact cost, token counts, and cache-hit rate. One refinement worth adding: Claude Code uses a second, smaller model for routine background work, and you can point it at something cheap — or free: export ANTHROPIC_SMALL_FAST_MODEL = glm-4.7-flash-free That puts K3 on the hard reasoning and a $0 model on the housekeeping. Kimi K3 in Cursor Cursor routes its chat / plan panel (Cmd/Ctrl + L) through a custom OpenAI-compatible endpoint. Setup: Open Cursor Settings → Models Add your LLM Gateway key under OpenAI API Key Enable Override OpenAI Base URL and set it to https://api.llmgateway.io/v1 Add kimi-k3 as a custom model and select it Be aware of the boundary: Cursor's Composer, inline edit (Cmd/Ctrl + K), and autocomplete are locked to Cursor's own backend and will not route through any external endpoint. Plan and chat with K3's full 1M context in Cursor; if you want K3 driving the actual agent loop, use Claude Code or Cline instead. Kimi K3 in Cline Cline is the straightforward one — it's built to bring y

Kimi K3 is the first open-weight model that makes the Kimi K3 vs Claude Opus 4.8 question worth asking seriously. On the Artificial Analysis Intelligence Index, K3 ranks fourth of 189 models — level with Opus 4.8 and GPT-5.5, behind only Claude Fable 5 and GPT-5.6 Sol. It also costs 40% less per token, and its weights are expected to be released by July 27. The honest answer is not "K3 wins" or "Opus wins" — it depends on what you're optimizing for. Here are the numbers, then the verdict. Benchmarks: Kimi K3 vs Claude Opus 4.8 Benchmark Kimi K3 Claude Opus 4.8 Artificial Analysis Intelligence Index 57 (4th/189) statistical tie GPQA Diamond 93.5% 93.6% Terminal-Bench 2.1 88.3%* 74.6%* SWE-bench Verified not published 88.6% Arena Frontend Code 1st (1,679) ranked below *Reported on different harnesses (Opus 4.8's score uses the Terminus-2 harness), so treat the Terminal-Bench gap as directional, not exact. Three takeaways. On graduate-level reasoning (GPQA Diamond) the models are statistically tied. On agentic terminal work, K3's published number is well ahead, with the harness caveat above. And in Arena's blind Frontend Code testing, developers ranked K3 first outright — ahead of every model, including Anthropic's. Where Opus 4.8 keeps an edge: SWE-bench Verified at 88.6% is a published, battle-tested result K3 has no counterpart for yet, and Anthropic's models remain the default target for agent harnesses — Claude Code, MCP tooling, and most agentic scaffolds are tuned against Claude first. K3 is three days old; its ecosystem is not. Pricing: 40% cheaper across the board Per-token rates through LLM Gateway (each provider's published pricing): Per million tokens Kimi K3 Claude Opus 4.8 Input $3.00 $5.00 Cached input $0.30 $0.50 Output $15.00 $25.00 Both have a 1M-token context window. K3's output limit defaults to 131K tokens but is configurable up to the full 1M in a single response; Opus 4.8 caps output at 128K. Concrete math: a coding-agent workload of 100M input a

Moonshot AI released Kimi K3 on July 16, and the benchmarks put an open-weight model next to the best closed ones for the first time. The catch is access. K3 sits on Moonshot's platform, GLM-5.2 on Z.ai's, DeepSeek V4 Pro on DeepSeek's, MiniMax M3 on MiniMax's — four accounts, four billing relationships, four API dashboards, all before you have written a line of code. LLM Gateway routes every one of them through a single OpenAI-compatible endpoint. One key, one bill, and a switch between Kimi K3 and any of 200+ models is a one-word change to your request. What is Kimi K3? Kimi K3 is Moonshot AI's flagship model for long-horizon coding and agentic work. At 2.8 trillion parameters — a mixture-of-experts design that activates 16 of its 896 experts per token — it is the largest open-weight model announced to date. Moonshot has committed to publishing the full weights by July 27, 2026. The specs that matter in practice: 1M-token context window (1,048,576 tokens), with output configurable up to the same 1M — enough to hold a large repository plus its docs in a single request Always-on reasoning — K3 thinks before every answer; there is no non-thinking mode Vision, tool calls, and JSON output supported out of the box Prompt caching at a 90% discount on repeated input Early results back up the size. K3 ranks fourth of 189 models on the Artificial Analysis Intelligence Index — level with Claude Opus 4.8 and GPT-5.5 — and took first place in Arena's Frontend Code evaluation in blind developer testing. It posted 93.5% on GPQA Diamond and 88.3% on Terminal-Bench 2.1, the strongest open-weight results published on both at release. Kimi K3 pricing Through LLM Gateway you pay Moonshot's published per-token rates: Tokens Price per million Input $3.00 Cached input $0.30 Output $15.00 The cached-input rate is the number to watch. Coding agents re-send the same system prompt, file context, and conversation history on every step, so in a long agent session most of your input tokens are

How Normal Software Engineers Actually Use AI in Their Daily Work Let's cut through the hype. You're not building the next AGI. You're a working software engineer with deadlines, legacy code, and a backlog that never shrinks. So how do you actually use AI tools in your day-to-day work? After surveying hundreds of developers and reflecting on real-world usage patterns, here's what normal software engineers are doing with AI—no Silicon Valley theatrics required. The Mundane But Invaluable: Code Completion and Boilerplate The most common use case is the least sexy: letting AI handle repetitive code. GitHub Copilot, Cursor, and similar tools excel at generating boilerplate that you'd otherwise copy-paste from Stack Overflow or previous projects. Real example: Writing CRUD endpoints in Express/TypeScript: typescript // Type this comment and let AI complete: // Create a REST endpoint for user registration with email validation app.post('/api/users/register', async (req: Request, res: Response) => { try { const { email, password, name } = req.body; // Email validation const emailRegex = /^[^\s@]+@[^\s@]+\.[^\s@]+$/; if (!emailRegex.test(email)) { return res.status(400).json({ error: 'Invalid email format' }); } // Check if user exists const existingUser = await User.findOne({ email }); if (existingUser) { return res.status(409).json({ error: 'User already exists' }); } // Hash password and create user const hashedPassword = await bcrypt.hash(password, 10); const user = await User.create({ email, password: hashedPassword, name }); res.status(201).json({ userId: user.id, email: user.email }); } catch (error) { res.status(500).json({ error: 'Internal server error' }); } }); Did AI write perfect code? No. But it gave you scaffolding to refine, saving 10-15 minutes of typing. That's the real win. The Game-Changer: Explaining Legacy Code and Obscure APIs Every developer inherits someone else's mess. AI tools shine when deciphering undocumented code or unfamiliar libraries. Pract

The first Lego X-Files set costs $200 and will be available in August.

I've spent my career helping people build software. This is a series of letters about what happens when the tools for building change faster than the principles behind building. Each one is a reminder that while the technology changes quickly, the questions that matter often stay the same. -- Dear past Jenna, The thing that drew you to tech in the first place, that it's always changing, is the thing that will keep you here. Tools change (sometimes for the better and sometimes not) almost weekly. People who never consider themselves technical, much less a developer, will build apps in an afternoon with this new programming language called English. Products will go from idea to deployed before you finish your first cup of coffee (you have a toddler now, so you rarely get the full cup before it goes cold anyways). "I don't know how to code" or "I'm not technical" is no longer a barrier to building. And that's exciting, given you've focused nearly your entire career helping others build software. But we can't confuse the ability to build with the wisdom to build. One of the most valuable habits you've developed over the last two decades is asking a simple question: "Should we build it?" For most of your career, "Can we build it?" was a hard question. Time, budget, complexity, maybe the tech wasn't there yet. But those constraints were usually temporary. With enough people, time, and money, just about anything is possible. But the real questions were always: Should we build this? Is this the best use of our time? Does this solve a problem our customers actually need solved, or does it create new problems? What are we choosing not to build? Those questions haven't changed, but the environment around them has. Today, almost anyone can build software. Between tools like Lovable, Bolt, Replit, Claude Code, Codex, and whatever's next, the barrier to building software is lower than it's ever been. Now the question "Can we build it?" is too easy to answer. It's almost always ye

Treasury Secretary Scott Bessent said the U.S. could sanction Chinese open AI models over alleged IP theft, expanding the Trump administration's campaign to slow China's AI advances.

Spotify's DJ-inspired tools are loads of fun, but probably won't land you any bar mitzvah gigs.

Google is launching Gemini 3.6 Flash alongside a new security model dedicated to quickly finding and patching security vulnerabilities. In a blog post on Tuesday, Google describes Gemini 3.5 Flash Cyber as a "cost-efficient and highly capable alternative" to larger, more expensive AI systems, such as the one offered by Anthropic's Mythos. The cybersecurity model […]

Learn more about Paramount's planned acquisition of Warner Bros. Discovery — a historic Hollywood megadeal valued at $111 billion — as it continues to develop.

It's important to be aware of disc rot, but there's no need to be afraid.

Deezer said more than 90,000 AI-generated tracks were uploaded daily on the platform in June.

I first slipped on Halliday's original smart glasses at CES 2025. I was not a fan. The glasses had a tiny, movable display window embedded into the frame that was incredibly finicky to see, and my 30-minute demo left me with achy eyeballs. It was an interesting concept with terrible execution, especially compared to the […]

Nintendo does a lot of weird things, but Splatoon might just be one of the strangest - or at least the boldest. The series started as an attempt to muscle into the ultra-competitive online shooter space, but in a way that feels a world away from the violence and bombast of Call of Duty or […]

Building AI Agents That Don't Hallucinate: Structured Workflows, Guardrails, and Per-Step Evaluation How we replaced fragile prompt chains with typed schemas, validation gates, and evaluation at every step — 94% task success vs 60% baseline The Prompt Chain Trap January 2024. We built a "research agent" — 12 prompts chained together: Decompose question → 2. Search planning → 3. Execute searches → 4. Extract facts → 5. Synthesize → 6. Fact-check → 7. Format → ... It worked 60% of the time. The other 40%: Step 3 returned malformed JSON → Step 4 crashed Step 5 hallucinated citations → Step 6 missed it Step 7 output wrong format → Downstream consumer failed No visibility into which step failed Debugging meant reading 12 LLM calls' worth of logs. Adding a step broke three others. The Shift: Agents as Typed Workflows We moved from prompt chains to structured workflows with: Pydantic schemas for every step input/output Guardrails that validate and auto-retry Explicit state machine (not implicit chaining) Evaluation harness per step (not just end-to-end) ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │ Decompose │──▶│ Search │──▶│ Extract │──▶│ Synthesize │ │ Question │ │ Planning │ │ Facts │ │ Answer │ │ │ │ │ │ │ │ │ │ In: Query │ │ In: Plan │ │ In: Results │ │ In: Facts │ │ Out: SubQ[] │ │ Out: Steps │ │ Out: Fact[] │ │ Out: Answer │ └──────┬──────┘ └──────┬──────┘ └──────┬──────┘ └──────┬──────┘ │ │ │ │ ▼ ▼ ▼ ▼ [Schema] [Schema] [Schema] [Schema] [Guardrail] [Guardrail] [Guardrail] [Guardrail] [Eval: 0.9] [Eval: 0.85] [Eval: 0.9] [Eval: 0.95] Core Abstractions # agent_eval/schemas.py from pydantic import BaseModel , Field from typing import Literal , Any class DecomposeInput ( BaseModel ): user_query : str context : dict = Field ( default_factory = dict ) class DecomposeOutput ( BaseModel ): sub_questions : list [ str ] = Field ( min_length = 1 , max_length = 5 ) requires_tools : bool reasoning : str class PlanInput ( BaseModel ): sub_questions : list [

AI agents no longer just answer questions — they book meetings, approve refunds, call APIs, update records, and chain together dozens of small decisions into a single outcome. Most of the time, this works quietly and well. But when something goes wrong, a troubling question surfaces: why did the agent do that? For a large share of deployed systems today, there's no good answer. The reasoning, the data consulted, the tools invoked, and the intermediate steps simply weren't recorded. This is the problem of undocumented agent decisions — and it's becoming one of the central risks of the agentic AI era. What "Undocumented" Actually Means An undocumented decision isn't necessarily a bad one. It's simply one that can't be reconstructed after the fact. In practice, this shows up in a few common ways: Output without reasoning. The system logs what the agent did, but not the chain of thought, tool calls, or data sources that led there. Ephemeral intermediate state. Multi-step agent chains often discard the "scratch work" between steps — the very material that would explain a decision — once the final output is produced. Reviewer blind spots. A human approves a final recommendation without ever seeing the reasoning that produced it, which looks like oversight but isn't meaningful oversight. No durable storage. Logs exist for a few days or weeks, then age out — so when someone asks for the record months later, it's gone. The common thread is a gap between acting and accounting for the action. This Is Becoming Urgent A few forces are converging to make this problem harder to ignore: Agents are doing more, autonomously. As agents move from single-turn assistants to systems that independently call APIs, touch databases, and trigger downstream workflows, the number of undocumented micro-decisions multiplies. A single customer request might now involve a dozen internal steps, each a potential decision point. Incidents are already happening. Surveys of enterprise AI deployments in 2

AI coding tools have gotten very good at one thing: generating code fast. What they haven't gotten good at is discipline. They don't know your architecture. They don't remember that you rejected a pattern last sprint. They don't know which parts of your context window are signal and which are noise. And they have no concept of a development workflow — no phases, no review gates, no verification steps. You describe what you want, they generate, and you hope the output fits. For small tasks this works fine. For anything that touches your real codebase at scale — refactors, new features with cross-cutting concerns, compliance-sensitive changes — the lack of workflow structure creates subtle, expensive problems that compound over time. We've been building Ortho to address two of these problems: workflow discipline (through ASES, a 6-phase AI development methodology) and context discipline (through a 9-component token optimization pipeline). This post explains both. The workflow problem with AI coding tools When a junior engineer joins your team, you don't just hand them a task and say "generate." There's a process: understand the codebase, plan the change, get the architecture reviewed, build it, test it, verify it, get it reviewed. The process exists because individual steps catch different categories of mistakes. AI coding tools collapse all of that into one step. You prompt. You get code. Done. The result isn't always bad code per file. The problem is architectural: the AI has no model of your layer boundaries, so it imports from layers it shouldn't touch. It has no memory of past decisions, so it re-proposes patterns you've already rejected. It has no verification step, so it confidently generates code that looks right but has subtle issues a reviewer would have caught in 30 seconds. The missing piece isn't a smarter model. It's a workflow. ASES: A 6-phase workflow for AI-assisted development ASES (v1.2) is the methodology built into Ortho's orchestration layer. It

It’s a lot : According to information obtained by The Tech , MIT is spending over $3 million on more than 500 AI surveillance cameras in academic buildings, residence halls, and outdoor areas along Memorial Drive. Installation of the new cameras, along with the wiring and infrastructure that will support them, began November 2025 and will likely continue until September 2026. Technical specifications for the cameras suggest that they will be capable of collecting real-time face and object classification data, including detection of motion, loitering, crowds, face masks, and camera tampering. Individuals can also be automatically classified on the basis of clothing color, gender, and age, up to a distance of 35 feet (11 meters) from the camera. According to a statement from MIT spokesperson Kimberly Allen, any collected data is “retained up to 30 days,” unless an exception is granted...

Bhavuk Jain discusses translating foundational AI into scalable mobile products. He shares the engineering challenges behind AI Wallpapers and Circle to Search, detailing how to implement robust runtime guardrails, fine-tuning, and seamless OS integration. For engineering leaders, he explains balancing UX constraints with model latency and infrastructure cost to deliver safe, reliable AI. By Bhavuk Jain

Gritt is coming out of stealth with $34 million and plan to automate the hardest tasks on construction sites.