Workshop your ideas into reality, in the Cloud and on your Desktop
London
Joined April 2007
- Tweets2.8K
- Following656
- Followers324
- Likes194
Pinned Tweet
Cool idea @MaziyarPanahi @googlegemma
Here is a demo you can play with
Gemma 4 + Sam 3.1 running all on a Mac via MLX
Build with @WorkshopAI
Repo 👇
Trying out a Bret Victor’s mini Dynamicland setup
chilang retweeted
New in Workshop: Auto Mode, powered by @RaylineAI.
Workshop already gives you access to frontier, open-source, and local/on-device models. But the real challenge is knowing which model provides the best cost/quality ratio for which sub-task.
With Auto Mode, you don’t have to choose manually.
In a single conversation, Workshop can use many different models together — routing each step to the model that is best fit for purpose across quality, speed, and cost.
That routing layer is Rayline.
You can now select Auto Mode directly from the model picker in Workshop.
And if you want to learn more about Rayline as a standalone product, check it out at rayline [.] ai
chilang retweeted
We open-sourced a local gateway for Claude Code.
The goal is simple: keep the Claude Code experience you already know, but use it with the models you want.
With the open-source gateway, you can connect Claude Code to other models using your own provider keys, whether that’s directly with model providers, through something like OpenRouter, or with local models running on your machine.
This is the more configurable version of what Rayline managed gives you out of the box: smarter model selection while preserving the coding-agent workflow you already like.
Rayline managed handles routing + inference for you.
The open-source gateway gives you control and lets you bring your own model connections.
GitHub repo link in thread.
[Example video below: Claude Code running with hybrid cloud + on-device AI. The main agent runs Opus in the cloud and orchestrates an Explore subagent that Rayline Local routes to a model running fully on device (Qwen3.6-35B-A3B, Q4), seamlessly, in a single session.]
chilang retweeted
Our current favourite model on the price/performance curve. Now upgraded.
GLM-5.2 by @Zai_org: the new open-source flagship with 1M context, Max-effort reasoning, and even sharper coding taste, still at a fraction of frontier cost.
Build with it now in Workshop.
Turn your ideas into reality 📷
chilang retweeted
You shouldn't need to replace Claude Code to make Claude Code 60–90% cheaper.
That’s why we built @RaylineAI.
Rayline is a model router that plugs into the harness you already use and routes each subtask to the best model capable of completing it successfully.
Models like Opus and GPT 5.5 can stay focused on orchestration, complex reasoning, and long-range planning, while faster and less expensive open-source or on-device models handle the rest.
Rayline can route across frontier, open-source, and on-device models. It abstracts away the complexity of knowing which model to use, when to use it, or what your hardware can support.
All it takes is a one-line change on your Claude Code settings. The result is the experience and quality you already know and love, at a fraction of the API cost.
This grew directly out of what we learned building @WorkshopAI. One of the things users value most is the ability to work across models from different providers. But we kept hearing the same question: can I use this intelligence inside the tools I already use?
Rayline is our answer.
Its ML router optimizes for the lowest cost of a successful outcome, not simply the cheapest model. Because a cheap model that fails, retries, or spins its wheels isn’t actually cheap.
Frontier models are incredibly valuable. They just shouldn’t be doing all the grunt work.
Rayline is available today at Rayline [.] ai.
Claude Code first. Codex next!
chilang retweeted
.@Benioff makes the pitch for @RaylineAI flawlessly.
Every part is there ...
- intelligent model routing,
- routing across cloud and on-device,
- the immediate focus on coding but the broader need beyond ...
Added some callouts for clarity :)
chilang retweeted
Claude Code's API rates are 8-10x more expensive than its subscription rates.
Most of those tokens are wasted spend.
We built @RaylineAI to fix that.
Rayline is a model router built specifically for Claude Code. You plug it in, and subagents get routed to open source and on-device models. Use Opus as your main agent overseeing grunt work delegated to open models. Quality holds. Costs drop 60-90%.
The key insight: model routing belongs at the API layer, not the harness layer. Harnesses are converging. LLMs are becoming interchangeable. The control point is the gateway, and that's where enterprises need to manage spend without forcing teams to adopt new tools.
What makes Rayline different from other routers:
- Built specifically for Claude Code
- Routes at the subagent/subtask level
- On-device routing via MLX (Qwen 3.6 and others)
- Built-in ML router trained for Claude Code tasks
We've been thinking about this problem for a long time. We built model routing into @WorkshopAI months ago. Others launched similar things last week.
Rayline is the first of its kind: a router built from the ground up for the coding agents you already use.
We're already routing billions of tokens per day for individuals and publicly traded companies.
Try it at Rayline[.]ai
(p.s. Codex is next!)
chilang retweeted
This is the exact thesis behind @RaylineAI
Route subagents to open models (and on-device), while letting users choose Opus/GPT 5.5 (or any of the "open frontier" models) for the main agent.
This type of model routing needs to live at the LLM gateway API layer ... use whatever harness you want, and whatever model you want
Good take
My guess is
- demand for intelligence is near infinite
- but 80% of workloads will be running on 99% cheaper models within 12-18 months
- 20% of workloads will still run on latest gen models where IQ maxing is important (scientific breakthroughs, higher level ochestrator agents?)
- rough analogy might be what % of macbooks or gaming PCs sold have the maxed out specs for CPU/GPU, prices are falling much faster than Moore's law here though
- this leads me to think the limiting factor will be energy and compute, not better models
At Coinbase we're working hard on routing prompts to cheaper models where appropriate, and in some cases have been able to keep costs roughly flat, while token usage continues to grow exponentially.
chilang retweeted
Gemini 3.5 Flash is now in Workshop.
Free for every paid plan today only!
Unmetered until 11:59 PM UTC ⏳
Go build something amazing!
🤖 Made with AI
chilang retweeted
Gemini is free on Workshop for Max subscribers from May 12–20!
Use Gemini for apps, workflows, internal tools, dashboards, agents, image tools, and more.
Available for both existing and new Max subs.
chilang retweeted
Want to win an election? Change your name.
Live data from London's #electiontresult2026 shows candidates higher in the ballot get more votes than their party colleagues in the same ward 72% of the time.
17 candidates so far that missed out on a seat:
ballot-order.workshop.build
chilang retweeted
Prompt-to-app is table stakes.
The hard part is what comes next: backend, data, APIs, auth, services, jobs, infrastructure.
Workshop is for building real software.
chilang retweeted
❤️
you can also download and use Gemma 4 models for subagents in @WorkshopAI.
A favorite setup of mine is to use Gemini 3.1 Pro for the main agent and Gemma 4 31b for subagents on my M4 Max.
chilang retweeted
Want to build with AI for free?
With local models in Workshop, you can build websites, dashboards, internal tools, workflows, prototypes, and more.
That means:
- Zero API costs
- Offline access
- Full privacy
Try local models today in Workshop Desktop.
chilang retweeted
You don’t have to use one model (or one provider!) for everything.
With Workshop, you can combine frontier and local models in the same workflow.
For example: Opus can be the main agent, and delegate specific tasks to Gemma 4 via subagents.
Better quality where it matters. Better privacy, speed, and cost where it counts.
One workflow, best model for each task.