Jay Nailer

Efoiling founder on a tiny island.
Building AI-native energy software from the middle of the Indian Ocean.

[Jay Nailer @jaynailer]

I'll stop writing posts.
Motion typing only.

[Jay Nailer @jaynailer]

@levelsio "Free"

But yes, I enjoyed the Fable ride too

[Jay Nailer @jaynailer]

@Amank1412 I never saw them as a giant..

Still use them for all old Google queries

[Jay Nailer @jaynailer]

@rom1trs Nice one!

Motion

[Jay Nailer @jaynailer]

@JakeFleshner Save energy

[Jay Nailer @jaynailer]

Sonnet 5 - Still confused on pricing.

Sonnet 5 is out, Fable will be back, and I’m still not clear: how cheap is Sonnet 5 actually?

The pricing details feel murky and I keep seeing conflicting takes. At this point… fuck it. I’m just going to switch between Opus and Fable.

[Jay Nailer @jaynailer]

@sflorimm I'm waiting for the China chips

[K2S @k2sbhai]

nvidia is casually giving you access to 5 frontier chinese AI models for free 😳

no credit card
no subscriptions
just one API key that unlocks everything

what you get for $0:

- DeepSeek V4 Flash for ultra-fast inference
- MiniMax M3 as a drop-in coding assistant
- Qwen3.5-397B for advanced reasoning tasks
- Kimi K2.6 for agentic workflows and long chains
- GLM 5.1 as a reliable everyday model

why this is huge:

> no paying separate subscriptions for different models
> no changing your existing workflows or tools
> no vendor lock-in since everything is OpenAI-compatible

getting started takes less than 2 minutes:

1. go to https://t.co/q50rSatNbb
2. sign up and verify your account
3. generate your nvapi key
4. set your base URL to https://t.co/92kkbFSf8D
5. pick any model and start building

supported models:

> minimaxai/minimax-m3
> qwen/qwen3.5-397b-a17b
> moonshotai/kimi-k2.6
> zhipuai/glm-5.1
> deepseek/deepseek-v4-flash

pro tip:

use DeepSeek V4 Flash for speed, Qwen for hard reasoning, Kimi for agents, and MiniMax as your daily coding companion

the best part?

one free key gives you access to 100+ models across NVIDIA's catalog

~40 requests per minute is more than enough for most developers and personal projects

5 frontier models that compete with GPT and Claude, all without spending a dollar

bookmark this and claim your free API key before the limits change 👀

[Jay Nailer @jaynailer]

@sporadica ha ha, true.

The last significant update I notices was the audio book feature. But that was ages ago...

[Jay Nailer @jaynailer]

@Bencera For me as a small business I really hope the cost will come down quite fast because I don't care if it's an American or a Chinese model or service provider. I think there will be different solutions for corporate, medium, and small-size companies and for the private end-user.

[Jay Nailer @jaynailer]

@nia_thinks I'm with porkbun since years.

They are doing their job

[Jay Nailer @jaynailer]

@MarioNawfal Yes, that's hot.

But an AC should be standard in France

[Jay Nailer @jaynailer]

Something for @levelsio

I am hope they never get that stupid here in Mauriritus...

[Jay Nailer @jaynailer]

Ok, Obsidian + Claude is nice

I tried working in Obsidian some time ago,
but couldn't get warm with it.

But with Claude and automatic rules for an LLM Wiki a la Kasperski it is great.

Instant full context for new chats, new projects about my company, brands, and digital life

[Jay Nailer @jaynailer]

I did the @levelsio move and migrated to Hetzner.

Before: AWS ~200USD monthly
Now: Hetzner ~25€monthly

Nice!

[Jay Nailer @jaynailer]

@wulfsagemakes @levelsio @mitsuhiko I will give it try!

[Jay Nailer @jaynailer]

@wulfsagemakes @levelsio @mitsuhiko I want switch from lastpass, the Proton package is on my favorite list.

I need a new/ additional VPN anyways.

So, you can recommend it?

[Jay Nailer @jaynailer]

Thoughts on Opus 4.8?

Looks good. I played a little bit around with it.

But in Cursor I still stick to GPT 5.5 and Composer 2.5

[Claude @claudeai]

Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.

Available today at the same price.

[Jay Nailer @jaynailer]

My work feels so much more structured and focus on actual impact to the usabililty of the app.

And the user gets feedback that their issues are handled and can feel the velocity of product improvements.

And will only gets faster...

[Jay Nailer @jaynailer]

Moved from intuitive vibe coding to LLM-driven product improvement:

[Jay Nailer @jaynailer]

- create feater/bug recommendations
- categorise them (prompt, skill, code base change)
- prioritize them
- on click: give cursor prompt to implement
- will inform user when feature got shippen/ fixed