We decided to run a small experiment: what if a group of experienced players built the best possible @fantasy_top_ Bronze deck together, with zero budget constraints?
We started by mapping out the strongest cards by star rating
1* and 2* were straightforward - @0xMakesy and @Crypto_McKenna are obvious picks
4* and 6* were the weak spots, with only a few decent choices (@JasonYanowitz and @a1lon9 for 4*; @waleswoosh, @trading_axe, @Pentosh1 for 6*)
But we had plenty of strong 3*, 5* and 7* options - including some completely cracked 5*s
That led us to two main structures:
– 1-2-3-5-7
– 1-2-5-5-5
Community picks:
3* – @HugoMartingale (then @zacxbt with @0xMerp as a risky wildcard)
5* – @Chilearmy123, @mert, @notthreadguy
7* – @DipWheeler (@MacroCRG is close; @fejau_inc was benched last minute)
Final drafts:
Deck 1: Makesy – McKenna – Hugo – Clemente – DipWheeler
Deck 2: Makesy – McKenna – Clemente – Mert – Threadguy
And one pick with great potential from me:
Makesy – McKenna – Zac – WaleMoka – Axe
Now we’ll see how far pure theory can take us in Bronze
GL, chat!
Nemotron 3 Ultra is fast and genuinely good
Compared it with 3 frontier models: DeepSeek V4, MiniMax M3, and Qwen 3.7 Max on 2 prompts
very impressive results
@gmoneyNFT@the_etherean yeah i believe ds truly can save a lot of money if we compared it to very expensive claude api for enterprise use on certain tasks but in this case i'd be concerned about privacy (and u still free to use oauth for personal use)
@gmoneyNFT@the_etherean of reasoning and burns a lot of tokens even on tasks that aren’t particularly complex
and on the tl I see a lot of positive reviews about it but they are ALL exclusively about how many tokens people used for very little money lol
@gmoneyNFT@the_etherean in short it can be used as an executor for trivial tasks but you can’t trust it with anything more complex or important. It may find bugs that don’t actually exist has a very high hallucination rate and in reality it’s not even that cheap, because it uses very long chains
Recap of Bronze Theorycrafting #36/55 @fantasy_top_@intern did well - top 2 2*, 627 pts. Not typical for him though - i was ready for either 0 or 900+
@WazzCrypto was the real star of the deck - 704 pts is great for a 3*, even if it was only top 2 (Pentoshi took top 1). @j0hnwang landed top 4, which is solid, but just 526 pts
@CharlotteFang77 wasn’t the best 4*, but met expectations - 665 pts, top 2.
@mert just delivered his base (kek) - 843 pts, top 6 7*
3365 pts = Top 25-49 = 0.015 ETH + 190 frags. Solid deck overall
All theorycraft results so far in the comments
Last Thursday, the Fantasy Top $100,000 jackpot was won! An incredibly awesome event where you had to assemble a lineup of 5 influencers who would score the most points during the tournament week.
But more than 4 days have passed, and we’ve gotten zero winner announcements, no posts about it from the official @fantasy_top_ account. On the site, we can still register our decks for the jackpot, even though that shouldn’t be happening.
I recently spoke with the winner @cryptonoobyN1, there are no announcements or notifications on the site about his win either, and naturally, he hasn’t received the money. Looks like they just didn’t plan this far ahead. 🫣
Only the Discord moderators are responding, thank you guys. The team itself has disappeared and isn’t answering. I honestly don’t know what to think.
Guys, are you even alive over there? @travisbickle0x@0xKipit@0xMikado
хей codex team
крутая обнова с /goal и /side, но я нашел неприятный баг - когда лимиты заканчиваются, кодекс продолжает бесконечно слать запросы и получать ошибку о том, что лимиты закончились, повторяет это каждые несколько секунд ralph loop of errors lol
я бы попросил кодекса разобраться с этим как обычно, но я уже почти сухой на этой неделе