# k3 Published articles for k3. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Kimi K3.1 Pro now has more ZDR providers on AI Gateway DevFeed: [Kimi K3.1 Pro now has more ZDR providers on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-11-04T17:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI Chat](), [context window](), [litellm](), [Chat completions endpoints](), [vercel ai sdk]() Tags: [3d-modeling](), [agent](), [agents](), [ai-gateway](), [ai-sdk](), [anthropic](), [api](), [apis](), [context-window](), [cost](), [failover](), [financial](), [image](), [input](), [install](), [interactive](), [k3](), [kimi](), [kimi-k3-1-pro](), [providers]() ### AI overview Kimi K3.1 Pro and K3.1 Pro Highspeed are available through AI Gateway from additional ZDR providers; Moonshot AI had previously been the only provider serving them. Both models accept image input and have a 1M-token context window. The text highlights coding, 3D modeling, and financial analysis, and says Highspeed is suited to interactive work where response time matters. They can be called through the AI SDK, Chat Completions, Responses, Anthropic Messages, or connected coding agents. ZDR can be required per request with the gateway provider option or for all team requests through dashboard settings. When enabled, routing is limited to providers with a ZDR agreement, and requests fail if none is available for the model. ### Source excerpt Kimi K3.1 Pro and its higher-speed version, Kimi K3.1 Pro Highspeed, are now available on AI Gateway from more Zero Data Retention (ZDR) providers. Until today, Moonshot AI was the only provider serving either model. K3.1 Pro improves on earlier Kimi models in coding, 3D modeling, and financial analysis, and can carry long agent runs from a small fix to a full project. Both models have a 1M-token context window and accept image input, and Highspeed suits interactive work like coding agents, where response time matters. Call either model through the AI SDK, the Chat Completions, Responses, and Anthropic Messages APIs, or a coding agent connected to AI Gateway. For Highspeed, use moonshotai/kimi-k3.1-pro-highspeed in place of the model name below. Use moonshotai/kimi-k3.1-pro as the model name: Setting zeroDataRetention in the gateway provider options requires ZDR for that request. To require it for every request your team sends, including requests from coding agents, turn on ZDR in the AI Gateway dashboard settings. With either one on, AI Gateway routes only to providers with a ZDR agreement. If none is available for the model, the request fails instead of going to a provider without one. To use it in Claude Code, Codex, Cursor, and more, install the latest Vercel CLI and run setup: Then select moonshotai/kimi-k3.1-pro or moonshotai/kimi-k3.1-pro-highspeed in the agent. See the coding agents guide for details. Try Kimi K3.1 Pro in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. You can view all language models available on AI Gateway. Read more ## Introducing India-based inference for Kimi K3 on Amazon Bedrock DevFeed: [Introducing India-based inference for Kimi K3 on Amazon Bedrock]() Original publisher: [Read original article]() Author: Naomi Teng Published: 2026-11-03T16:32:54Z Content type: tutorial Language: en Sources: [Artificial Intelligence]() Topics: [LLM optimization](), [amazon](), [AI Chat](), [Language Engineering](), [Image](), [Amazon CloudWatch]() Tags: [amazon](), [amazon-bedrock](), [announcements](), [based](), [bedrock](), [code](), [country](), [cross](), [example](), [global](), [india](), [indic](), [inference](), [intermediate-200](), [k3](), [kimi](), [kimi-k3](), [language](), [mumbai](), [offers](), [options](), [post]() ### AI overview Amazon Bedrock's `in.moonshotai.kimi-k3` inference profile routes Kimi K3 processing to India Regions, supporting workloads with India data residency requirements. The excerpt describes invoking it through Mumbai and using the model's image input to read Indic-language documents and answer Hinglish questions without a separate OCR or translation service. Example workflows can extract schema-defined JSON, normalize dates and digits, transliterate names, summarize documents, and handle follow-up questions. The article recommends high image detail when small text matters, with increased input-token use, and advises validating extraction results for the intended use. It also covers cross-Region routing, optional prompt caching and request service tiers, and monitoring; invocation logs and metrics are stored in the source Region. ### Source excerpt Amazon Bedrock now offers in-country inference in India for Kimi K3, keeping processing within the Mumbai and Hyderabad Regions. This post explains the global and geographic cross-Region inference options and shares example code for reading and understanding Indic-language documents with Kimi K3. ## Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed are available now in AI Gateway DevFeed: [Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed are available now in AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-10-23T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI Infrastructure](), [AI-assisted coding](), [vercel ai sdk](), [Chat completions endpoints](), [Claude](), [AI Chat](), [Finance]() Tags: [3d](), [3d-modeling](), [agent](), [agentic](), [agentic-tasks](), [agents](), [ai](), [ai-gateway](), [ai-sdk](), [analysis](), [animation](), [anthropic](), [api](), [api-keys](), [apis](), [gateway](), [k3](), [kimi](), [kimi-k3](), [kimi-k3-1-pro]() ### AI overview Moonshot AI's Kimi K3.1 Pro and faster Highspeed variant are available through AI Gateway. The excerpt says K3.1 Pro improves coding, 3D modeling, financial analysis, and agentic tasks over Kimi K3, while retaining a 1M-token context window and image input. It describes coding work from codebase fixes to full projects and long agent runs, plus linked financial models and research grounded in source documents. Both variants can be called through the AI SDK, Chat Completions, Responses, Anthropic Messages, or supported coding agents; the excerpt gives `npx vercel ai-gateway setup` for configuring agents and distinct model names for each variant. AI Gateway offers usage and cost tracking, retries, failover, routing, reporting, Zero Data Retention support, and API-key budgets. It states that pricing matches providers without markup or an inference platform fee, including BYOK. Broader provider support is expected after Moonshot's planned model-weight release by November 4. ### Source excerpt Kimi K3.1 Pro from Moonshot AI is now available on AI Gateway, along with Kimi K3.1 Pro Highspeed, a faster version for interactive use. K3.1 Pro is the first model in Moonshot's K3.1 series. It builds on Kimi K3 with improvements in coding, 3D modeling, financial analysis, and agentic tasks, and keeps the 1M-token context window and image input. For coding, K3.1 Pro handles work ranging from small fixes in an existing codebase to building a full project, including long agent runs with many steps. It builds front-end pages and interactive scenes with stronger typography and animation than earlier Kimi models. In finance, it can carry a task from start to finish, such as building a linked financial model with adjustable assumptions and scenarios, or writing research grounded in source documents. K3.1 Pro makes fewer factual errors than earlier Kimi models on difficult prompts. Moonshot AI plans to release the model weights by November 4: support for more providers will follow. Call either model through the AI SDK, the Chat Completions, Responses, and Anthropic Messages APIs, or a coding agent connected to AI Gateway. For Highspeed, use moonshotai/kimi-k3.1-pro-highspeed in place of the model name below. Use moonshotai/kimi-k3.1-pro as the model name: To use it in Claude Code, Codex, Cursor, and more, install the latest Vercel CLI and run setup: Then select moonshotai/kimi-k3.1-pro or moonshotai/kimi-k3.1-pro-highspeed in the agent. See the coding agents guide for details. Try Kimi K3.1 Pro in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. You can view all la ## Get Started with Kimi K3 on CoreWeave Dedicated Inference DevFeed: [Get Started with Kimi K3 on CoreWeave Dedicated Inference]() Original publisher: [Read original article]() Author: CoreWeave Published: 2026-10-22T16:27:39Z Content type: tutorial Language: en Sources: [Coreweave Blog Posts]() Topics: [AI Infrastructure](), [model-serving](), [GB300 NVL72](), [vllm](), [Dynamo](), [GB200](), [LLM optimization](), [AI Factory]() Tags: [2](), [3](), [agentic](), [ai](), [ai-compute](), [all](), [content-type](), [coreweave](), [dedicated-inference](), [deploy](), [gb300](), [gb300-nvl72](), [guide](), [inference](), [k3](), [kimi](), [kimi-k3](), [nvidia](), [nvl72](), [this](), [with]() ### AI overview CoreWeave Dedicated Inference deploys customer-provided Kimi K3 weights on NVIDIA GB300 NVL72, with a tenant-isolated gateway, autoscaling, and vLLM serving on NVIDIA Dynamo. Setup involves staging Kimi K3 and a DSpark draft model in CoreWeave AI Object Storage, granting access, creating a gateway and deployment with a validated engine configuration, verifying the endpoint, and then cleaning up. Customers must supply a valid Kimi K3 license and weights; billing is per GPU-hour, with no ingress or egress fees. In CoreWeave's 16-GPU tests of multi-turn agentic workloads with a sustained 95% cache hit rate, GB300 scaled to concurrency 32 at 198,000 output tokens per minute per GPU. At concurrency 32, that was 209% higher throughput than GB200, with 133% higher p50 decode throughput; GB200 peaked at concurrency 16 and became unusable at 32 for latency-sensitive interactive workloads. The stack also supports KV-cache-aware routing, speculative decoding, and prefill/decode disaggregation. ### Source excerpt Follow this step-by-step guide to deploy Kimi K3 on CoreWeave Dedicated Inference on NVIDIA GB300 NVL72. ## Moving from Docker Compose to k3s: Flannel, Traefik and the Gateway API in simple terms DevFeed: [Moving from Docker Compose to k3s: Flannel, Traefik and the Gateway API in simple terms]() Original publisher: [Read original article]() Author: BillaCode Published: 2026-10-13T10:52:20Z Content type: article Language: en Sources: [DevOps.dev]() Topics: [Docker Compose](), [traefik]() Tags: [api](), [backend](), [bootstrap](), [docker-compose](), [flannel](), [gateway-api](), [go](), [helm](), [home-server](), [k3](), [k3s](), [kubernetes](), [microservices](), [open-source](), [production](), [queue](), [scripts](), [traefik](), [transcoder](), [video](), [workers]() ### AI overview The Kithly backend--a Go API, queue workers and a video transcoder--was moved from Docker Compose on a home server to k3s. The article says it will explain why k3s was chosen and clarify the roles of Flannel, Traefik and the Gateway API, then describe the cutover, including an eight-minute production outage. The supplied excerpt ends during a trimmed Compose configuration, so it does not establish the migration steps, how the networking and routing components were configured, or how the outage was resolved. ### Source excerpt Hi Guys, the Kithly backend (a Go API, a few queue workers and a video transcoder) has been running on a home server with Docker Compose... Continue reading on DevOps.dev " ## Kimi K3 is now available in GitHub Copilot DevFeed: [Kimi K3 is now available in GitHub Copilot]() Original publisher: [Read original article]() Author: Walker Chabbott Published: 2026-08-06T17:27:25Z Content type: release Language: en Sources: [GitHub Changelog]() Topics: [GitHub Copilot](), [GitHub Copilot CLI](), [Visual Studio Code]() Tags: [agentic](), [agentic-coding](), [ai](), [ai-model](), [billing](), [business](), [coding](), [copilot](), [enterprise](), [fireworks-ai](), [github](), [github-actions](), [github-copilot](), [incident](), [k3](), [kimi](), [kimi-k3](), [model](), [pricing](), [usage-based-billing]() ### AI overview Kimi K3, an open-weight model hosted by GitHub on Fireworks AI, is rolling out to Copilot Pro, Pro+, Max, Business, and Enterprise users across listed Copilot clients. Availability is gradual, and quality and performance will be monitored. The text gives conflicting rollout updates: an August 6 editor's note says rollout resumed after an earlier note announced a temporary pause. Usage-based billing is at provider list pricing; the earlier note specifies $3 per 1M input tokens, $15 per 1M output tokens, and $0.30 per 1M cached input tokens. For Business and Enterprise, Kimi K3 is off by default; administrators must enable its policy, and are advised to review open-weight models against their organization's security, compliance, and data-governance requirements. ### Source excerpt Editor's note (August 6, 2026): We have resumed rolling out Kimi K3 in GitHub Copilot. This model is billed at provider list pricing under usage-based billing. See GitHub Copilot's pricing... ## Under the Hood: Serving Kimi K3 DevFeed: [Under the Hood: Serving Kimi K3]() Original publisher: [Read original article]() Author: Shree Murthy Published: 2026-07-30T17:10:40Z Content type: article Language: en Sources: [DigitalOcean]() Topics: [LLM Techniques](), [AI Infrastructure](), [Multi-GPU](), [NVLink](), [model architecture](), [Distributed Systems & Parallel Computing]() Tags: [ai](), [amd](), [architecture](), [attention](), [b300](), [bandwidth](), [benchmarks](), [cache](), [capacity](), [deployment](), [digitalocean](), [distributed](), [engineering](), [experts](), [fabric](), [gpu](), [gpus](), [hardware](), [k3](), [kimi](), [kimi-k3]() ### AI overview DigitalOcean says serving Kimi K3 on launch day required 8-GPU NVIDIA HGX B300 or AMD Instinct MI350X servers: its roughly 1.56 TB of weights need about 195 GiB per GPU, leaving limited memory for KV cache and activations. Neither GPU can hold the full weights alone, so high-speed NVLink or Infinity Fabric interconnects support distributed attention and expert-parallel computation. DigitalOcean used llm-d for its native support of mixed GPU types, enabling onboarding to both NVIDIA and AMD platforms. The excerpt identifies hardware selection and stack integration as parts of the effort but ends mid-sentence, so it does not provide the full account or verification details. ### Source excerpt DigitalOcean launched Kimi K3 on day 0. It's already one of the most popular models on the platform and across the market: second most likes on Hugging Face, sixth most traffic on OpenCode. Getting a model this size running well on day zero took real work across several teams. Thanks to Moonshot AI, Inferact, RadixArk, NVIDIA, and AMD for the help getting there. Standing up a new model, integrating it into DigitalOcean's Inference Engine, and showcasing its unique attributes on day 0 takes three things: the right hardware, a tuned serving stack, and rigorous verification against Moonshot's own benchmarks. Here are the lessons we learned along the way: Hardware selection and implementation We selected NVIDIA HGX™ B300 and AMD Instinct™ MI350x GPUs to run K3 because these instances provide the memory capacity, FLOPs, and interconnect horsepower necessary for a model of K3's size and architecture. We built our distributed inference stack with llm-d because it includes native support for GPU type heterogeneity. This let us quickly onboard K3 to both AMD and NVIDIA platforms. Kimi K3 has roughly 2.78 trillion total parameters, 896 routed experts, and an attention stack that interleaves 69 Kimi Delta Attention (KDA) layers with 24 Gated Multi-head Latent Attention (MLA) layers. Kimi-K3 weights are ~1.56 TB in total, which requires about 195 GiB per GPU. Given such a large memory footprint for the weights alone, and a need to keep enough headroom for KV cache and activations, the practical unit of deployment is an 8x NVIDIA HGX B300 or AMD Instinct MI350X server. Both have 288GB of VRAM capacity, and after loading the weights, there is still some amount of practical memory left for the KV cache. Entire weights cannot be loaded on a single GPU. That's where the high-speed scaled-up NVIDIA's NVLink or AMD's Infinity Fabric is critical to ensure there is enough interconnect horsepower for bandwidth intensive, latency sensitive attention and expert parallel computations. Model ## The Dodo Digest: The Best Startup Ideas Usually Already Exist DevFeed: [The Dodo Digest: The Best Startup Ideas Usually Already Exist]() Original publisher: [Read original article]() Author: Rishabh Goel Published: 2026-07-24T00:00:00Z Content type: opinion Language: en Sources: [Dodo Payments Blog]() Topics: [Business, Industry, Usability, User Experience, User Research, Web Strategy](), [product analytics]() Tags: [account](), [ai](), [analytics](), [best](), [better](), [builder](), [categories](), [dodo](), [existing](), [ideas](), [inventing](), [k3](), [kimi](), [kimi-k3](), [making](), [multi-account](), [newsletter](), [products](), [spotlight](), [startup](), [v3]() ### AI overview The newsletter argues that startups can win by making familiar products cheaper, simpler, or more accessible rather than creating new categories. It presents Kimi K3 as an example, citing competitive performance alongside lower cost and open weights, while noting that country-by-country AI leadership is not its main takeaway. Dodo Payments' Analytics v3 brings real-time revenue, comparisons, country and payment metrics, retention, MRR, churn, and recovery into a customizable dashboard with shareable chart cards. Other updates include paid trials, trial misuse prevention, multi-account payouts, chargeable-currency pricing, and Google and GitHub login. The Scira AI spotlight says localized payment options helped serve customers in 25+ countries; it reports 3x revenue growth since switching providers and launching plans in under 24 hours. ### Source excerpt Kimi K3 shows why making existing products dramatically better beats inventing new categories. Plus Analytics v3, paid trials, multi-account payouts, and a Scira AI builder spotlight. ## How to code with Kimi K3 better than Fable and no restrictions DevFeed: [How to code with Kimi K3 better than Fable and no restrictions]() Original publisher: [Read original article]() Author: Pieter Levels Published: 2026-07-17T00:00:00Z Content type: tutorial Language: en Sources: [@levelsio (Pieter Levels)]() Topics: [codex](), [Claude Code](), [Jailbreak]() Tags: [account](), [ai](), [api-key](), [build](), [bypass](), [claude](), [claude-code](), [code](), [codex](), [fable](), [how-to](), [install](), [k3](), [kimi](), [kimi-k3](), [opencode](), [permissions](), [run](), [settings](), [tech]() ### AI overview The post gives a setup sequence for using Kimi Code with OpenCode: ask Codex or Claude Code to install OpenCode, create a Kimi account, pay for the stated $19/month membership, obtain an API key from kimi.com/code/console, then run OpenCode and use `/connect` to connect to Kimi Code with that key. The author recommends BUILD mode, selected with Shift+Tab, and says bypass permissions can be enabled in `/settings`. These are the post's instructions and preferences; it provides no comparison evidence for the title's claim about Fable. ### Source excerpt How to code with Kimi K3 (better than Fable and no restrictions) today: Ask Codex or Claude Code to install OpenCode (yes it will 😂!) Go to kimi.com, make an a...