George M

[George M @luckymoooon]

@gaeldelalleau @__smiz @Kog__AI @grok any plan to add at least 8B llms, qwen 27B support would be even better

[George M @luckymoooon]

@buildwithdjdev @StrengthClub1 Thank you, what's your opinion on delivering xcode sdk as portable wasm modules(or you know a better solution to reach portability) to avoid containers? This company is using cheerx tech to run virt x86 apps in browser https://t.co/Oh77K1LGpC
There is also https://t.co/cEne71V1Nu

[George M @luckymoooon]

@buildwithdjdev @StrengthClub1 It's working only on macs? Can you provide info how to config it in scenarios like this https://t.co/BDv6OLFZPb

[George M @luckymoooon]

@josevalim José, can you try riveter project from github, a combination from ralph loops and jj versioning created by rivet author? may be response to what you're looking for

[George M @luckymoooon]

@mitchellh @misterfitzie Hey Mitchell, did you see the mojo x8 performance against rust in a 50k loop object creation/destruction, written in february 24 on the mojo creator's website https://t.co/gbTEwzmXmD and that mojo reached v 1.0 beta last week, who knows, maybe rust is not the bun last destination

[George M @luckymoooon]

@buildwithdjdev wasn't the m4 max with 48 GB but 40 core GPU and 546 GB/s bandwidth a better option at 3000$ as seen on the apple refurbished store?

[George M @luckymoooon]

@joelteply @spiritbuun @emreceng_ Hey Joel, any opinion on the modular/max engine based on the mojo lang, in many scenarios is superior to cuda and other engines/sdks, also have support for apple silicon since september and rdna consumer gpus since a week or so

[George M @luckymoooon]

@fantparl @0xSero how they are supposed to have better bandwidth - will come with upcoming GDDR7 40 gbps or more than 512 pins?

[George M @luckymoooon]

@haknkskn @xgodil @Kimi_Moonshot what's best - cline, roo code or kilo code?

[George M @luckymoooon]

@xenonwellz @AmanVirk1 how about elysia, which one would you choose

[George M @luckymoooon]

@itsjustmarky @0xSero how fast is glm-4.7-flash running on strix-halo, did you run it?

[George M @luckymoooon]

@heiwen @saltyAom did you try tanstack start, how does it compare to rr7?

[George M @luckymoooon]

@TobiM can you check the glm-4.7 with the same promt to compare, the Pro plan, can be purchased at $144/year with 200m tokens every 5 hours

[George M @luckymoooon]

@ivanfioravanti @s_j_thapa the same thing for Ultra, apple need to double the bandwidth to compete with RTX PRO 6000 with 1.8 TB/s at 7000$ with 96 GB VRAM, not to mention risc-v cips like https://t.co/qb4oQfUvCx or https://t.co/LcFkmnQ6ok (with gpddr7 will have 1.3TB/s bandwidth, x2.5 more) at 1400$

[Jim Keller @jimkxa]

TT-Ascalon is officially IP released. Go build.
RiscV is now high performance.

Really happy with the team, quality of the release, quality of the support IP and DV infra.
Open source hardware is great.
https://t.co/rFV1miEiYO

[George M @luckymoooon]

@ivanfioravanti @s_j_thapa M4 Max already has 550 GB/s, there is no advantage over the upcoming (hope in Q3) AMD Medusa Halo with 384 pins and lpddr5x at 10700 Mt/s support (520 GB/s)
the actual Strix Point you can buy a minipc at 1500with 96 GB RAM, so apple has to double the bandwidth to be competitive

[George M @luckymoooon]

@qubitium did you hear about modular and their max engine running on mi325x, here a comparison vs nvidia h200 https://t.co/c3vhm1wmIO

[George M @luckymoooon]

@rdraculla @EffectTS_ @steipete can you give https://t.co/SDXIcLzYdZ a try, its author migrated from fp-ts to effect, then to pure ts for his library

[George M @luckymoooon]

@rdraculla @EffectTS_ @steipete got it, re effectTS, what framework did you use before it or still use?

[George M @luckymoooon]

@rdraculla @EffectTS_ thank you, but how do you get 6.4 billion tokens a month, is openai providing tokens at such big speed?

[George M @luckymoooon]

@alkimiadev @0xSero ok, re ide, zed is very performant and getting a lot of vibe but their worktrees support is not yet finished, compared to getpochi vsc addon, but vsc itself as an electron app is lazy, I find myself searching to what to migrate to, have to give the cli a new try