Guillermomo

@DankEngine13 @0xSero @chutes_ai The best models are often overloaded and slow unfortunately.

@0xSero Please look into @chutes_ai . It offers AI inference for a fraction of the price of many subscriptions, and you get most of the open source frontier models including TEE.

Best subscriptions on market based on your budget: 10$ a month: #1 - Opencode go #2 - GLM basic #3 - MiniMax basic #4 - Huggingface Pro 20$ a month: #1 - GPT-Plus #2 - Kimi-Code #3 - Mix and match from 10$ tier 50$ a month #1 - Qwen Code #2 - Mix and match from above

@megastuffs @jorgiting @0xSero It's a very fast version of Kimi though, great for agents like Hermes but not so great for coding in my experience.

@jorgiting @0xSero 28 a month and only kimi rn

@alspinu @0xSero Could you give a review of Ollama Pro? do you use it for Coding or Agent?

@0xSero I feel Ollama Pro is more generous than Kimi Code.

@builtbykern @0xSero Qwen is *ok*, definitely not worth $50.

@0xSero Why 50$ on qwen?

@GLun @0xSero Lite plan is gone unfortunately. Too good value for money I suppose.

@0xSero I suggest adding AlibabaCloud Lite to the $10 list

@northontorga @0xSero Went to sub today and it seems I can't make a personal account from the UK? Pretty odd. Might try a VPN.

@guillermode20 @0xSero was waiting byteplus release seed-2.0-pro on the coding plan… looks like they did 5d ago and i wasn’t aware… i just signed up (i’m in Brazil), will run some tests tomorrow and post my findings

@megastuffs @Cryprobuddha421 @0xSero Appreciate the insight. Shame it's not good for coding.

@guillermode20 @Cryprobuddha421 @0xSero Horrible if you are coding, tps is way lower than other providers and models like glm5 or kimi2.5 are usually overloaded. Limits are very generous though.

@EggrollOverIt @0xSero Definitely hold onto it. Great selection. Hopefully they keep it for the foreseeable for you.

@ura_soul @0xSero Can vouch for Alibaba, on the pricier side but crazy usage and great model selection.

@0xSero alibaba $50 tier is great. xai cheaper tier includes glm 5.1

@northontorga @0xSero Can you sub to byteplus from western countries (like the UK)? It seems like good value for money. Can you give a review?

@0xSero nice list, but you missed stepfun and byteplus buddy

@Cryprobuddha421 @0xSero I would love to hear about your experience with Chutes. Have you tried it as a coding agent?

@0xSero May I argue that Chutes.ai should be in the $10/month option? I run hermes agent on GLM5 and the $10 a month covers ALL my API calls. It is also private decentralized and gives you access to the best open-source models for cheaper than the competition.

@EggrollOverIt @0xSero You can't sub to it anymore, it's gone unless you already had a sub.

@0xSero what about the Alibaba Coding Plan Lite (legacy) - $10

@xEvoRaiJiN @0xSero Barely any usage.

@0xSero Why not Claude code for 20$ a month?

@munichaser @0xSero Fire Pass is great, well worth grabbing while it's cheap. Kimi K2.5 Turbo is fast as heck.

@0xSero what about fireworks fire pass?

@hlitoroy @0xSero That deal is long gone, and they just iced their $10 a month lite plan. The $50 plan is pretty great though, crazy usage and good model selection.

@0xSero $3 a month: alibaba cloud

@0xSero You're toon based? That's awesome. I'm there for uni and it's a lovely city, reckon I'll stay.

Who’s in London? I would love to meet a few people over the next week. All my friends are in Newcastle ); won’t be able to meet them this week

@0xSero I would love it :) On the very cusp of making my 100% rust game clipping software public. Far lower CPU and memory usage than apps like Medal and completely open. The newest GPT models have been instrumental for optimisation.

Do you want 3 months of Codex Pro? Comment under the post we are selecting a 5 people very soon 3 MONTHS of Pro. I genuinely think this is the highest leverage sub we have

@bridgebench But a model that makes up just under a third of the time is? Seems like a pretty petty criticism when the frontier are all within punching distance. That <3% isn't going to make a huge difference between 5.4 and 5.1.

GLM 5.1 improved on hallucination over GLM 5. Still not enough. GLM 5 fabricated 38.6% of the time. GLM 5.1 brought it down to 34.1%. Better. But still the worst frontier-class model on BridgeBench Hallucination. Z.ai closed the intelligence gap with Claude Opus 4.6 and GPT 5.4 on coding benchmarks. They haven't closed the trust gap. A model that makes things up a third of the time isn't ready for production codebases. bridgebench.ai

@NosytLabs @pelaseyed You meen OpenCode?

Announcing Grok CLI v1 🔥 There are many coding agents. This is Grok’s. ✅ All the features the others have. All of them. ✅ Grok-native search (x/web) + multi-agent ✅ Remote control from Telegram ✅ Hackable and 100% Open Source (MIT) npm i -g grok-dev Github ⏬

@turtlekiosk Yeah, Rust is sick. Making a Medal clone at the minute, managed to get the CPU and RAM usage ridiculously low

im kinda convinced the thing to do now is learn rust