stosssik

Original approach for their annual shareholder decks. I liek that.

Can't remember if the correct parameter for your LLM is max_tokens or max_completion_tokens ?

Say no more: we are launching modelparams . dev, the first complete model parameters database, open and collaborative.

You can browse it (UI), query it (API) or types in your app (NPM).

A curated list of LLM APIs with permanent free tiers, no trial credits, no credit card traps.

Info included: rate limits, max context, and supported modalities.

Here's the list per provider:

Cohere (https://cohere.com/)

  • Command A (111B)
  • Command R+
  • Command R
  • + 3 more model (https://docs.cohere.com/docs/models)

Google Gemini (https://ai.google.dev/)

  • Gemini 2.5 Flash
  • Gemini 2.5 Flash-Lite

Mistral AI (https://mistral.ai/)

  • Mistral Small 4
  • Mistral Medium 3
  • Mistral Large 3
  • + 3 more model (https://docs.mistral.ai/getting-started/models/models_overview/)

Z AI (Zhipu AI) (https://z.ai/)

  • GLM-4.7-Flash
  • GLM-4.5-Flash
  • GLM-4.6V-Flash

Inference providers - Third-party platforms that host open-weight models from various sources.

Cerebras (https://cerebras.ai/)

  • llama3.1-8b
  • gpt-oss-120b
  • qwen-3-235b-a22b-instruct-2507
  • zai-glm-4.7

Cloudflare Workers AI (https://developers.cloudflare.com/workers-ai/)

  • @cf/meta/llama-3.3-70b-instruct-fp8-fast
  • @cf/meta/llama-3.1-8b-instruct-fp8-fast
  • @cf/meta/llama-3.2-11b-vision-instruct
  • + 5 more models (https://developers.cloudflare.com/workers-ai/models/)

GitHub Models (https://github.com/marketplace/models)

  • gpt-4.1
  • gpt-4.1-mini
  • gpt-4o
  • + 7 more models (https://github.com/marketplace/models)

Groq (https://groq.com/)

  • llama-3.3-70b-versatile
  • llama-3.1-8b-instant
  • llama-4-scout-17b-16e-instruct
  • + 7 more models (https://console.groq.com/docs/models)

Hugging Face (https://huggingface.co/)

  • Meta-Llama-3.1-8B-Instruct
  • Mistral-7B-Instruct-v0.3
  • Mixtral-8x7B-Instruct-v0.1
  • Phi-3.5-mini-instruct
  • Qwen2.5-7B-Instruct

Kilo Code (https://kilocode.ai/)

  • bytedance-seed/dola-seed-2.0-pro:free - Modality: Text | Rate Limit: ~200 req/hr
  • x-ai/grok-code-fast-1:optimized:free - Modality: Text (code) | Rate Limit: ~200 req/hr
  • nvidia/nemotron-3-super-120b-a12b:free
  • arcee-ai/trinity-large-thinking:free - Modality: Text (reasoning) | Rate Limit: ~200 req/hr
  • openrouter/free - Modality: Text | Rate Limit: ~200 req/hr

LLM7.io (https://llm7.io/)

  • deepseek-r1-0528 - Modality: Text (reasoning) | Rate Limit: 30 RPM (120 with token)
  • deepseek-v3-0324 - Modality: Text | Rate Limit: 30 RPM (120 with token)
  • gemini-2.5-flash-lite - Modality: Text + Vision | Rate Limit: 30 RPM (120 with token)
  • + 3 more model (https://llm7.io/)

NVIDIA NIM (https://build.nvidia.com/)

  • deepseek-ai/deepseek-r1
  • nvidia/llama-3.1-nemotron-ultra-253b-v1
  • nvidia/nemotron-3-super-120b-a12b
  • + 3 more models (https://build.nvidia.com/models)

Ollama Cloud (https://ollama.com/cloud)

  • llama3.1:cloud
  • deepseek-r1:cloud
  • qwen2.5:cloud
  • gemma2:cloud
  • mistral:cloud

OpenRouter (https://openrouter.ai/)

  • deepseek/deepseek-r1-0528:free
  • deepseek/deepseek-chat-v3-0324:free
  • qwen/qwen3.6-plus:free
  • + 9 more free models (https://openrouter.ai/models?q=free)

SiliconFlow (https://siliconflow.com/)

  • Qwen/Qwen3-8B
  • deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
  • deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
  • + 3 more model (https://siliconflow.com/models)

RPM = requests per minute • RPD = requests per day. TPM = Tokens per minute • TPD = Tokens per day • RPS = Requests per second • All endpoints are OpenAI SDK-compatible.

I've talked to over a hundred OpenClaw users over the past two months. Cost comes up in almost every conversation. People set up their agent, use it normally, and end up with bills they didn't expect. $141 overnight from a misconfigured heartbeat. $800 in a month on a multi-agent setup. $30 burned doing barely anything.

The article digs into why this happens and what you can do about it. The core problem is that without optimization, every request hits your most expensive model, your system context loads on every call, and your conversation history grows with each exchange. It adds up fast.

The fixes range from simple config changes to architectural decisions. Routing tasks to the right model instead of sending everything to Opus. Using skills instead of spinning up multiple agents. Leveraging prompt caching on the provider side. Keeping your context lean. Running local models for lightweight tasks. And tracking costs daily instead of discovering a surprise bill at the end of the month.

Two deployments documented 77% and 80% cost reductions through these approaches. All sources and community reports are linked at the bottom. Happy to answer questions.

I mapped the major players building the OpenClaw ecosystem, just 2 months after its release.

What's unfolding around OpenClaw is unlike anything I've witnessed in open-source AI.

In 60 days: - 230K+ GitHub stars - 116K+ Discord members - ClawCon touring globally (SF, Berlin, Tokyo...) - A dedicated startup validation platform (TrustMRR) - And an entire ecosystem of companies, tools and integrations forming around a single open-source project.

Managed hosting, LLM routing, security layers, agent social networks, skill marketplaces.

New categories are taking shape in real time.

Some of these players are only weeks old. And established companies like OpenRouter, LiteLLM or VirusTotal are shipping native integrations.

Whether you're a VC exploring AI infra, an operator running agents, or a founder building in this space, this is the landscape right now.

Some of these startups are already pulling real revenue. Alternatives are stacking thousands of GitHub stars on their own. OpenRouter recently raised funding. The money and the users are already here. Most of what's on this map didn't exist 60 days ago.

This is what happens when an open-source project launches with the right building blocks at the right moment.

This resonates a lot. And we’re working on something in the same space: a way to build MCP aps for non technical people. If there are builders here who like experimenting, we’re looking for beta testers: -> https://manifest.build

Totally agree. Vibe coding will generate lots of internal AI apps, but turning them into reliable, secure, governed services still requires real engineering, which is exactly why we’re building https://manifest.build. It lets non-technical teams build Agentic apps fast through an AI powered workflow builder while giving engineering and IT a single platform to add governance, security, data access, and keep everything production-ready at scale.

Thanks for the perspective. Could you be more concrete about what specifically doesn’t change with vibe-coded apps? Have you recurring friction points in mind that force a handoff to professional engineers once these apps need to scale ?

The Scouring is an indie real-time strategy game with action-RPG elements built on a custom C++ engine. It features destructible environments, day-night cycles that affect gameplay, and a unique mechanic allowing players to delegate army control to AI while focusing on hero character development.

SQLite is a powerful, lightweight database that's widely deployed across mobile devices and applications. Despite being perceived as a beginner tool, it offers ACID compliance, can scale to 100TB, requires no separate server process, and supports advanced features like Write-Ahead Logging (WAL) for improved concurrent performance. The article explores SQLite's core features, demonstrates its durability and transaction handling, and introduces LibSQL, a distributed fork designed for modern cloud applications that addresses traditional SQLite's concurrency limitations.

Looks super interesting. I hadn't heard of Graphiti before, but the idea of giving Cursor some kind of persistent, structured memory across sessions definitely sounds useful.

A promising approach—especially if it proves as simple and low-cost at scale. It’s obviously not going to "disrupt" the plastics industry overnight, but it could offer a valuable local alternative, particularly in regions dealing with massive plastic waste imports. The real question is whether this kind of tech can evolve outside of patent lock-in and centralized exploitation models.

I doubt this is a “new era” for TC. At this point, it’s more of a brand than an impactful media outlet. Maybe Regent will just squeeze out whatever SEO juice is left.

Really like your vision. The “no endless feeds, no humblebrags, just real people” pitch really resonates.

LinkedIn feels exhausting, so I’m definitely going to give OpenSpot a try.

Wishing you the best

This is fantastic — simple, fun, and with just the right vibe for exploring ideas quickly. I love how focused it is: no clutter, just straight into the groove.

Do you see this staying as a passion project, or are you considering a business model at some point ?

Good catch! You're absolutely right. We switched from SHA3 to bcrypt a few weeks ago, but forgot to update the docs.

Thanks a lot for pointing it out. I just created an issue to track this, and we’ll fix it shortly: https://github.com/mnfst/manifest/issues/361

Hi, Manifest founder here.

PocketBase is great too. Manifest brings something different: it's fully code-based, so you can stay in your IDE, use AI tools like Copilot or Cursor to build your backend, and keep everything versioned in Git. It fits naturally into dev workflows, and we’ve got more coming soon on the AI side, both around AI and new features.

Hey HN,

Over the past few months, Manifest has sparked a lot of interest, and an engaged community has formed around it. Until now, it wasn’t really possible to run a Manifest backend in production—but that’s changing. With 4.10, we’re opening a new chapter: for the first time, we might start seeing Manifest-powered backends going live!

What is Manifest?

Manifest is a a AI-friendly micro-backend that powers websites and apps

Manifest consists to have a complete backend that fits into a single YAML file, easy to read and modify.

From this YAML file, you’re creating All the logic, The data, And the storage.

What’s new in 4.10?

- Postgres support – Now works with PostgreSQL - Middleware support – Extend functionality as needed - S3 storage compatibility – Handle file uploads seamlessly - Default values – Automatically populate missing fields - Improved password security – Switched to bcryptjs

This is a big step for Manifest, but we know there’s still a long way to go. What do you think? What blockers do you see for using this in production?

We’d love to hear your feedback—positive, negative, critical, all of it! If you’ve been following Manifest, this is the moment where things get real.

Discussion & full details: https://github.com/mnfst/manifest/discussions/352 Check out the repo: https://github.com/mnfst/manifest

My colleague and I recognize the current complexity of backend development, where even basic functionalities require mastering a myriad of technologies. This complexity underscores the urgent need for more streamlined approaches.

To address this, we've developed a new methodology specifically designed to simplify backend creation for frontend developers. To refine this approach and gain insights into its potential impact, we conducted a survey to understand frontend developers' perspectives on backend development challenges.

Give us your insights about the other side of the screen !

Complete this 3-minute survey, and we'll donate €3 to Techfugees, an NGO dedicated to assisting refugees through technology.

Who are we ? Seb and Bruno are 2 project's holders committed to building a free and open-source backend solution.

This questionnaire is part of our research for a project aimed at understanding current challenges in development workflows, specifically focusing on backend creation and deployment.

Your answers will guide the initial direction of our project, which is committed to being open source.

The findings will be shared by the end of February to inform our product development and help address prevalent pain points in the tech community.

Thank you

Hey Devs! Got 3 minutes to share your backend development experiences?

Participate in our quick survey, and we'll donate 3€ to Techfugees, an awesome NGO empowering refugees through tech.

Your insights will help us shape a free, open-source backend solution that truly meets your needs .

can you list the features that are lacking in django?

we want to develop a simple invoicing web app as fast as possible for our business.

For that motive, we are going to use CASE, a simple tool to quickly generate dashboards and add our own custom logic.

This simple guide will follow 4 steps:

1. Install CASE 2. Generate entities 3. Add properties 4. Generate a PDF on invoice creation

Using CASE.app, this developer build an AI-Powered NFT Generator.