Alex’s AI posts

Prices & limitsOpenAIAlex23:39

And so the fairy tale comes to an end

Starting in December, ChatGPT web chat will no longer be free and unlimited. Just like Claude, chatting will now eat into your usage limits.

Prices & limitsOpenAIAlex17:13

Codex is doubling usage limits across all Pro plans!

The $100 plan is now 10x. Sam Altman decided not to cut the $200 plan in half, so it stays at 20x, which is pretty awesome too. And the $500 plan is…

ReleasesGoogleAlex18:15

Gemini 4 Carbon is out

According to synthetic benchmarks, it’s now the best AI model in the world. The Artificial Analysis Intelligence Index seems to confirm that. But…

Tools & updatesOpenAIAlex08:50

Apple Mail and Calendar plugins have landed in GPT

Also, we've got another reset ✅

NotesOpenAIAlex18:14

The new Terra is less capable than Sol

From what I understand, OpenAI is positioning it as a cheaper model focused on computer use.

ReleasesOpenAIAlex17:56

GPT-6.1 Terra is out!

Prices & limitsOpenAIAlex08:14

Tibo is resetting Codex limits for the fourth time in the past five days

The last time we saw this kind of generosity was back in July, when we got six resets in six days.

Prices & limitsOpenAIAlex09:27

Tibo reset the Codex limits while everyone was asleep

As they say, don’t look a gift horse in the mouth. But I wouldn’t mind burning through a few more tokens.

NotesAnthropicAlex07:36

Thariq is leaving Anthropic

I think he was the strongest public face of Claude Code, arguably even more than Boris Cherny. He wasn’t just shipping features and announcing…

Tests & comparisonsMoonshot AIAlex18:48

Open-weight model Kimi K3.1 Pro ranks second on the Artificial Analysis leaderboard

Tests & comparisonsMoonshot AIAlex17:27

Kimi's benchmarks are absolutely mind-blowing

Is there a new sheriff in town?

Tests & comparisonsMoonshot AIAlex17:20

Kimi K3.1 Pro benchmarks are out!

Of all these benchmarks, I trust Terminal-Bench 4.0 the most. And the results are impressive: Kimi K3.1 Pro scores 65.7, just 0.7 points behind the…

ReleasesMoonshot AIAlex16:30

Kimi K3.1 is out now!

Available outside China via Kimi Code. Pricing per 1M tokens: Input / Output: ~$2 / ~$10 I’ll post the benchmarks as soon as they’re available.

Prices & limitsOpenAIAlex08:00

Tibo is resetting Codex limits for the second day in a row

Looks like the compute bottleneck is finally a thing of the past. Resets are great, of course, and I’ve got plenty of ideas on how to use them.…

Prices & limitsOpenAIAlex08:40

Another Codex reset

You’ve got until this evening. But I’d recommend using it up beforehand.

LeaksAnthropicAlex19:24

Claude Duet, a proactive AI agent from Anthropic, is coming soon

OpenAI has Dots, xAI has Grok Bot, and there’s also Hermes Agent. Now Anthropic is joining the race. Looks like we’re moving from the era of…

Prices & limitsOpenAIAlex08:29

While OpenAI are prepping the new Astra and Terra, they dropped another reset

LeaksOpenAIAlex09:53

OpenAI’s API has revealed three new models:

• aurora-green-alpha • aurora-blue-alpha • aurora-red-alpha About an hour later, three new models appeared on Design Arena: aspen, cypress, and…

LeaksOpenAIAlex09:16

OpenAI API has revealed three new models:

• aurora-green-alpha • aurora-blue-alpha • aurora-red-alpha These appear to be new cybersecurity models that could replace Daybreak Blue and…

Prices & limitsOpenAIAlex17:10

Codex reset confirmed

You've got a couple of hours to use up your limits

LeaksOpenAIAlex18:49

Looks like Astra 6.1 is coming

OpenAI rolled out a private test model gpt-6-astra-api-priv-ev3

Prices & limitsOpenAIAlex10:26

Day 10 of the OpenAI sprint

No new ideas, so a reset

Tools & updatesAnthropicAlex22:24

Claude Code just added idle compaction, and this could save a LOT of usage

Imagine you have a session with 500K tokens of context. You come back 90 minutes later, your prompt cache has expired, and Claude has to process…

Tests & comparisonsGoogleAlex16:12

Gemini 4 Argon benchmarks are out

We've actually seen these benchmarks before, back when the model was first announced. Looks really strong in text, vision, etc. thanks to Google's…

ReleasesGoogleAlex15:49

Gemini 4 Argon is out!

Only Ultra users for now

LeaksAnthropicAlex13:51

Fable 5.5 appeared in Azure’s East US 2 model catalog with GenerallyAvailable status

https://github.com/BerriAI/litellm/commit/7dd53d6adc6213e84f62f1366ba721d09f4af992

Prices & limitsOpenAIAlex04:14

Whined my way into a Codex reset, keeping the team on their toes even on a public holiday

Prices & limitsAAnysphereAlex23:50

Elon Musk just combined Cursor, Grok, and Twitter subscriptions into one

Now, for $8/month, you get X Premium, Cursor, and Grok all in one. And for $30/month, you also get Grok Bot. If you’ve been paying for Cursor and…

ReleasesMetaAlex19:34

Meta just released Muse Wonder

According to the benchmarks, it's right up there with GPT 6 Astra and Claude Opus 5.5. The results look impressive, but the model might be heavily…

Tools & updatesOpenAIAlex15:53

OpenAI sprint week 1 recap

Day 1: WIN Sped up Sol and Astra from ~30 to ~50 tokens per second. Most people use exactly these models, so it's a clear win. Day 2: NOT ACCEPTED…