Legit

3.3K posts

Legit banner
Legit

Legit

@legit_api

api guy with little-to-no wisdom

Katılım Ekim 2018
903 Takip Edilen14K Takipçiler
Sabitlenmiş Tweet
Legit
Legit@legit_api·
Made by Veo 3 "A squirrel attempts to launch a homemade rocket into space"
English
20
36
495
188.3K
Legit retweetledi
M1
M1@M1Astra·
Early preparations for Claude Opus 5 have been spotted again. Seems imminent.
English
34
30
614
264.7K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Kimi K3 ranks 3rd on VoxelBench just 100+ Elo points behind Fable! and a huge uplift from K2.6 (28th)
Voxelbench tweet media
English
5
17
189
33.9K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Kimi K3 lands on VoxelBench The outputs look great so far! Full benchmark run still ongoing ⏳️
Voxelbench tweet media
English
1
3
54
1.9K
Legit retweetledi
sankalp
sankalp@dejavucoder·
dario after kimi k3 drop
Lietuvių
18
71
1.8K
47.7K
Legit retweetledi
Theo - t3.gg
Theo - t3.gg@theo·
@MatthewBerman - Writes better code - Writes less code - Understands intent better - Right tuning of "eagerness" (I find 5.6 to be a bit too eager) - Smarter overall
English
60
6
588
24.6K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Claude Fable 5 max "Om Nom playing a level in Cut the Rope"
Voxelbench tweet media
English
1
2
20
1.2K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
gpt-5.6 sol on max effort It can "write" really nice in voxels
Voxelbench tweet media
English
0
3
38
1.6K
Legit retweetledi
MarsForTech
MarsForTech@MarsForTech·
Seedance 2.5 (30s, single take)
English
40
32
375
128.8K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
GPT-5.6 Sol overtakes Fable 5 on VoxelBench the highest scoring model yet with over 1k votes!
Voxelbench tweet media
English
4
8
160
6.5K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
A first impression of GPT-5.6-Sol (Max) I think the picture speaks for itself.
Voxelbench tweet media
English
0
9
81
7.1K
Legit retweetledi
Mark Zuckerberg
Mark Zuckerberg@finkd·
(1) Today we're releasing Muse Spark 1.1 -- a strong agentic and coding model at a very low price. It's available through our new Meta Model API and in Meta AI.
English
5.3K
3.6K
45.6K
23.3M
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Grok 4.5 hit the leaderboard! Impressive step forward from @SpaceXAI!
Voxelbench tweet media
English
2
4
47
4.2K
Legit retweetledi
Angel 🌼
Angel 🌼@Angaisb_·
New strings found in Claude suggest Opus could get its own separate usage limits, apart from the shared ones First they take Fable from us, now they want to limit Opus even more Come on Anthropic...
Angel 🌼 tweet media
English
17
4
120
9.6K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
SpaceXAI has released Grok-4.5 and it is a huge improvement over previous Grok versions. It matches Opus-4.8 in benchmarks. Visit us at voxelbench to see more outputs from Grok 4.5 or vote to shape the leaderboard! For Elon, we ran 2 Mars related prompts.🚀
Voxelbench tweet mediaVoxelbench tweet media
English
1
2
48
5.3K
Legit
Legit@legit_api·
@cheatyyyy @nadzi_mouad yea it should stay available on the API it will also remain in claude web, but with added cost now through usage credits :/
English
0
0
1
67
Legit
Legit@legit_api·
@nadzi_mouad idk, probably but Anthropic has shown they can serve the model perfectly fine for Claude Max users so that whole thing about it returning "when capacity allows" is sus; uptime on claude web for Fable 5 is at an ATH this month, there is more than enough capacity rn!
English
2
0
1
109
Mouad
Mouad@nadzi_mouad·
will it end today at midnight? @legit_api
English
1
0
2
124
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Claude Fable 5 (Max) > A Rabbit in a dress shirt and blazer pointing at the minute hand of a pocket watch.
Voxelbench tweet media
English
1
4
41
2.4K
Legit retweetledi
Voxelbench
Voxelbench@voxelbench·
Claude Fable 5 on Max effort > Atlas and P-body making a dramatic pose
Voxelbench tweet media
English
1
8
38
3.2K