Cristian 🇦🇷🇷🇺🇵🇸

A favor del anti-imperialismo y de la revolución de los neo-esclavos-asalariados. 🇦🇷🇷🇺🇵🇸

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@Valent14F Mucha injerencia?? Vos decís??
https://t.co/mtGTZxUjvO

[Mati Morro @matias_dapipon]

@beneluckx @crisjs_19 @MundoEConflicto

[Sandy Cecilia @Saceci]

Lo que Estados Unidos No quiere que sepas.
🇺🇸🇮🇱

Este es el video por el cual le eliminaron la cuenta de Tik tok a Diego Saúl Reyna.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@Mustaineista @MundoEConflicto Hay pobrecitos ellos que les echan la culpa de todo...

https://t.co/5VDFSmqGL1

[Yes to books, No to guns 🏳️‍🌈🌎🌍🛬 @CUPCAKEWILLIAM6]

@9mmScorpion

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@nothingto12 it's probably just because they're in a different business. But it still piques my curiosity.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@nothingto12 Yes, that makes total sense. But my point is, if on top that they offer something similar to, let's say, Ollama ($20 --> $60), that would mean some extra millions that are not going to the competence anymore.

I'm sure they have a reason for not offering it, and, as you said,

[Alexey Fateev @superalesha]

Fuck you, @ArtificialAnlys

Your new index says Claude Haiku 5.5 is smarter than DeepSeek V4.1 Flash 1017 and GLM-5.3.

Haiku is Anthropic's small cheap model. DeepSeek 1017 is the model Cursor put on its cost frontier this month, and Cursor says no model scores higher at those prices. I run it every day on my 4x 3090. I dont believe for a second that Haiku beats it.

Look at how they got there. 60% of the index now runs on private tests nobody outside can see. GDPval-AA, SciCode and the long context test are gone. In comes a private coding benchmark from Cognition. And the open weights gap goes from 3 points to 12 in 1 release, 4x. Released open models are now 26 points behind the top.

1 day before v5, on the old version, they gave Ling 3.1 Flash 1019 a 44. Now the best released open model gets 32.

The weights didnt change. The tests did, and you cant see the tests.

This is a bought benchmark. A leaderboard where you cant check a single question, that widens the gap 4x overnight and puts Haiku above DeepSeek, is marketing for closed labs. Its not measurement. I hope at least you were paid enough money to publish such garbage. Liars.

Publish the questions or stop calling it an index.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@nothingto12 Just to say thanks, dude! That's the exact comparison I was waiting for! 👏👏👏

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@DefesaSulGlobal "The Zionist regime that now holds the Argentine state"**, would be directly...

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@gonsanchezrey1 Estamos de acuerdo en que aún hay que ser cautelosos, no? Y guiarnos por acciones/resultados mas que por personalidades.
https://t.co/5yhQ2CxOOg

[Puflito @Puflito___]

- "Ay que bueno que ganó Lula, le puso un freno al sionismo en América Latina"

- Lula:

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@JovenPangolin @armesillaconde No es tratado de defensa mutua, es de coperación.
Pero vamos, no es algo menor... muy turbio de todas formas. Gracias por compartirlo.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@nothingto12 @CommandCodeAI Bro, please keep including Command Code. Right now they're one of the big players in the scene, and the goal isn't to advrtise or bash them; its just to get the truth about their service quality.
Sorry for them if doesn't help

You're really helping the community with this, btw!

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@Menteenblanc01 @armesillaconde Bueno, no sabemos eh... El cuento de los bandos enfrentados es conocido, pero la ambición de estos tipo tampoco conoce límites.
Las internas de poder ocurren hasta dentro de las familias. Yo no lo descartaría aún.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@JovenPangolin @armesillaconde Acuerdo mutuo de defensa entre Brazil e Inglaterra !?
Wuau, esa si que es fuerte

[El Gordo Galpón @TavoCibreiro]

¡Histórico! En Francia, parte de la policía decidió dejar de reprimir y marchar con el pueblo.
Game Over, Macron.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@opencode New test.
To be fair, here it performed much better:
https://t.co/HtAi1PS6Fp

The price benefit you offer is worth that tiny difference in cache rate/throughput performance.

[Naz @nothingto12]

🧵 Round 1 - DeepSeek API vs OpenCode Go vs CommandCode GOAT

Finally here! This time, you're able to view the demos yourselves and judge.

https://t.co/VOjMdbIW7H

NOTE: The run for CommandCode GOAT failed on me after many retries. Still, it did produce working code at the end.

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@opencode Are we paying for quantized version models?
Please explain @opencode
https://t.co/k9bsDcxNrw

[Leo · Building w/ AI @leo114119]

Opencode 被网友扒出来 Deepseek v4.1 Flash 1017 的质量远不如官方，应该是量化后的自部署无疑了。。

作者的原话：Go 更便宜、更慢，还有超时和更差的质量。 每轮质量胜者都是 DeepSeek API，OpenCode Go 频繁超时。

[Cristian 🇦🇷🇷🇺🇵🇸 @crisjs_19]

@nothingto12 Yep, several are reporting this 🥴
https://t.co/k9bsDcxfBY

[Leo · Building w/ AI @leo114119]

Opencode 被网友扒出来 Deepseek v4.1 Flash 1017 的质量远不如官方，应该是量化后的自部署无疑了。。

作者的原话：Go 更便宜、更慢，还有超时和更差的质量。 每轮质量胜者都是 DeepSeek API，OpenCode Go 频繁超时。

[Rabbi Jeff @RabbiJeffx]

BREAKING: After being exposed for bankrolling politicians, JEWISH @AIPAC is changing tactics.

Instead of donating through its PAC, it’s recruiting individual donors to send cash DIRECTLY to politicians.

@grok Start with TRUE or FALSE.