Leo

1.5K posts

Leo banner
Leo

Leo

@AgenticLeo

Founder engineering autonomous AI systems | Obsessed with machine-to-machine communication | RM

Katılım Eylül 2021
901 Takip Edilen369 Takipçiler
Leo
Leo@AgenticLeo·
Luna (medium) is cheap and fast so quota hit is low, and it's not great but not bad at handling this
English
0
0
0
11
Leo
Leo@AgenticLeo·
I’ve started assigning a Luna subagent to maintain a live HTML status page for every long-running coding thread. Progress, tasks, blockers, review findings, etc. Some tasks/goals run for days, and they are hard to track by reading walls of text.
English
1
0
0
29
Leo
Leo@AgenticLeo·
@pigeon__s @testingcatalog Plus Sesame uses Gemma under the hood, which is miles behind, Live 1 in high effort is actually decent But worth a try imo
English
0
0
0
3
Leo
Leo@AgenticLeo·
@pigeon__s @testingcatalog Tbh Sesame is better and more natural, but not by that much, Live 1 is great and close. But lack of tool use is so annoying. With Sesame you can take notes and do some stuff at least. But I expect oai to add tools soon, so at that point yea, idk
English
1
0
0
5
🚨 AI News | TestingCatalog
🚨 AI News | TestingCatalog@testingcatalog·
Sesame is working on new Calendar and Email connectors for its iOS app. Users will be able to provide Sesame agent an access to read and potentially operate these apps. Besides that, there is an internal menu for testing custom skills where custom prompts can be added. It would be super useful if that feature will get released to the public too. Currently, Sesame is still a “go to” voice mode for me because it is both duplex but also has access to tools (reminders and note taking). I like GPT Live 1 as well but it lacks tool access at this moment.
🚨 AI News | TestingCatalog tweet media🚨 AI News | TestingCatalog tweet media🚨 AI News | TestingCatalog tweet media🚨 AI News | TestingCatalog tweet media
English
5
1
36
5K
Logan Kilpatrick
Logan Kilpatrick@OfficialLoganK·
I am hiring a TPM lead for @GoogleAIStudio, the job is to accelerate our progress in every way possible across product, GTM, engineering, and more. If you are AI pilled, high agency, and want to push the frontier in Google DeepMind, consider working with us : )
English
92
30
893
88.6K
Tibo
Tibo@thsottiaux·
Tomorrow might be 8M active user celebration day. Just saying
English
1.6K
397
11.8K
2.2M
Leo
Leo@AgenticLeo·
@AIatMeta @Meta @alexandr_wang @finkd WhatsApp connector when?! I want to connect LLMs to my messages like Slack or discors or the other sources! Third party slop feels off
English
0
0
1
65
Leo
Leo@AgenticLeo·
Luna is just as incredible as Sol, I replaced a bunch of workflows and did some evals and tests. It is simply incredible for the price. I want to see Luna intelligence get even cheaper! (And smarter)
English
0
0
1
57
Leo
Leo@AgenticLeo·
Even tho I nuked the ($200) Fable limits in ~2 days and I will only have 1 to day next week to use it
English
0
0
1
58
Leo
Leo@AgenticLeo·
As a sidenote. I am enjoying the product of both these insane token limits for Sol and extension of Fable
English
1
0
1
66
Leo
Leo@AgenticLeo·
This is possibly the best PR campaign oai has ever done. Though it must be pretty expensive for them, I bet it's making enough people hop the train to make it worth it
Tibo@thsottiaux

Added a banked reset to 500k users of ChatGPT Work and Codex. What’s been happening: - Just released the ability to use a banked reset from web and mobile. Before today you could only do it in the desktop app - Had an issue where less than 10% of users who used a banked reset and it didn’t actually reset. This was during a 2 hour window. - It was hard to find all the exact users where it didn’t apply, so instead granted a banked reset to everyone who pressed the reset button in that 2 hour window. This also gives us the opportunity to validate the new infra ahead of tomorrow Tomorrow we will celebrate our 7M active users milestone and grant the first banked reset across all of our ChatGPT Work and Codex users. We will also release a bunch of updates to the desktop app, addressing a ton of your feedback. Talk soon

English
2
0
17
1.9K
Leo
Leo@AgenticLeo·
@RobertJBye We pretty much all burned through our limits thinking it would get pulled today... :/
English
0
0
2
77
Leo
Leo@AgenticLeo·
@kimmonismus Thing is, Sol is comparable to Fable, so if Opus is below that it becomes "irrilevant" for the frontier. And likely pareto frontier too. So it's tricky
English
0
0
0
64
Leo
Leo@AgenticLeo·
@AndrewCurran_ Given the many reports that Fable still destroys Sol on design, I wonder how this is
English
0
0
6
163
Andrew Curran
Andrew Curran@AndrewCurran_·
Praise the Sun.
Andrew Curran tweet media
Design Arena@DesignArena

BREAKING - OFFICIAL RESULTS: GPT-5.6 Sol by @OpenAI is 1st overall on Design Arena with an Elo of 1353. This puts GPT-5.6 Sol above Claude Fable 5 by @AnthropicAI and in the same performance band as GLM 5.2 by @Zai_org on frontend design. This is an 18-position and 60-point Elo leap from GPT-5.5. GPT-5.6 Sol also establishes a new Pareto frontier for preference vs. speed, faster than any model at this performance. Congratulations to the @OpenAI team on the launch!

English
15
8
321
27.8K
Leo
Leo@AgenticLeo·
@claudeai Also a rate limit reset right?
English
0
0
0
38
Claude
Claude@claudeai·
We're extending Claude Fable 5 access on all paid plans, as well as keeping Claude Code’s weekly rate limits 50% higher, through July 19.
English
6.7K
7.3K
73.8K
26.3M
Leo
Leo@AgenticLeo·
@jxnlco Will we haver get super fast Codex transcription? I use it so much it's sometimes a bottleneck lol
English
0
0
0
331
jason
jason@jxnlco·
Import the Wispr Flow dictionary into Codex! Wispr Flow stores its learned dictionary in a local SQLite database on macOS: ```text ~/Library/Application Support/Wispr Flow/flow.sqlite ``` Just tell codex to make sure to merge them with the existing `dictationDictionary` field in config.toml
English
41
13
391
47.5K

Keşfet

@prd_008 @pigeon__s @testingcatalog @OfficialLoganK @GoogleAIStudio @thsottiaux @AIatMeta @Meta