That B Guy

@MiguelxCarlon Fellas is it gay to treat your woman good?

Every time I hear this take I think about every famine, drought, food shortage, or starvation event through history and think about how there is never a single account of a person who is upset about the weight they are gaining

therapists are borderline not allowed to give advice. if your therapist is giving you direct advice on how to handle a situation you genuinely need to leave asap

Farmers shouldn't allow their workers to use tractors and tools because that will just make the workers more expensive!

Surprise! Next month, I’m collaborating with my brother @UFO_FGC to host the Ego Chall Inaugural Tekken Tournament! 32 man cap - PC - $1K pot bonus! This will take place the Sunday after next months @IronFistMonthly Pull up! Reg now! start.gg/tournament/ego…

Happy Birthday to me, I'm only 29 years old and have done many amazing things in the FGC 😎 This year I gave myself the gift of home improvement, thanks for all of the support. I'll be live often this year while also traveling to an exclusive set of events and working on ICFC!

@mykeryu For some reason my post is getting flagged for inappropriate content? You all might want to look at your filters.

Yes, we are reading the threads. Thank you to everyone already participating. This portal is now the only place we are consistently reading, so if you want to be heard, this is the place.

The #TEKKEN Dev Feedback portal will launch along #TEKKEN8 Season 3! This portal is a hub for gathering information, bug reports and feedback regarding battle systems and gameplay. Access it here: Web🤜 tekken-dev-feedback.bn-ent.net Discord 🤜 discord.gg/ZF2jNbSQPg

Some of these changes in this latest Tekken patch just have no direction and it really feels like they changed things without any sort understanding of how the game works

Everyone's TL in about 2 hours

@BlackoutzUlt I'll be back in a week. LOOOOOOL

@thatbguy Lmaooo u shoulda known better

@BlackoutzUlt why he do me like that man? hahahahhaha

Best save ever

Whoever decided to make Goku flicker between his base and Super Saiyan 1 form understands Dragon ball at it’s core

Life goals

these are not contradictory statements

So happy I could make it out to @IronFistMonthly. Been trying for a long time but my schedule never let me. I manned up after doing a halfrox race. Exhausted as all hell, hungry, sleepy, and tired. Pulled up and made top 8 with a 7th place finish. Old man @TEKKEN #NCFGC

IRON FIST MONTHLY 40 Grateful to help build such a passionate, welcoming community here in Greensboro. What started as a shared love for fighting games has grown into something deeper, a space where people connect, support, and joke with each other. See y’all next month!

IRON FIST MONTHLY 40 TEKKEN 8 TOP 8 1. @_MHendo 2. @GenghisD0n 3. @AceUnlimited00 4. @JJBREEZE_ 5. @Static_T8 7. @IamJNY5 7. @thatbguy ggs to all! See y'all on April 18th!

GPT-5.4 loses 54% of its retrieval accuracy going from 256K to 1M tokens. Opus 4.6 loses 15%. Every major AI lab now claims a 1 million token context window. GPT-5.4 launched eight days ago with 1M. Gemini 3.1 Pro has had it. But the number on the spec sheet and the number that actually works are two very different things. This chart uses MRCR v2, OpenAI’s own benchmark. It hides 8 identical pieces of information across a massive conversation and asks the model to find a specific one. Basically a stress test for “can you actually find what you need in 750,000 words of text.” At 256K tokens, the models are close enough. Opus 4.6 scores 91.9%, Sonnet 4.6 hits 90.6%, GPT-5.4 sits at 79.3% (averaged across 128K to 256K, per the chart footnote). Scale to 1M and the curves blow apart. GPT-5.4 drops to 36.6%, finding the right answer about one in three times. Gemini 3.1 Pro falls to 25.9%. Opus 4.6 holds at 78.3%. Researchers call this “context rot.” Chroma tested 18 frontier models in 2025 and found every single one got worse as input length increased. Most models decay exponentially. Opus barely bends. Then there’s the pricing. Today’s announcement removes the long-context premium entirely. A 900K-token Opus 4.6 request now costs the same per-token rate as a 9K request, $5/$25 per million tokens. GPT-5.4 still charges 2x input and 1.5x output for anything over 272K tokens. So you pay more for a model that retrieves correctly about a third of the time at full context. For anyone building agents that run for hours, processing legal docs across hundreds of pages, or loading entire codebases into one session, the only number that matters is whether the model can actually find what you put in. At 1M tokens, that gap between these models just got very wide.