cheaty
@cheatyyyy
24, sde @ stealth startup, performative nerd
891 Following 4.1K Followers
Sol and Opus today, haven't had a double release like this in a while, going to be very explosive and exciting
For everyone wondering it’s GPT 6 Sol today - before Aeon.
Claude Opus 5 is currently ridiculously fast in Claude Code (and yes, it's not the stealth Opus 5.5 test)
either they spun up a lot more compute right before launch and are testing it or a lot more people are currently being routed to Opus 5.5 while Opus 5 sits without load
Show more
can somebody at Anthropic/Claude fix the TUI rendering entirely breaking when having Hindi/Devanagari script being pasted into Claude Code
just a single character of the script it entirely breaks rendering
just let me paste my meeting transcript for Claude to get work done 🙏
Show more
You are not ready for the taste and design uplift coming from the soon to be released Anthropic models
A lot of domains like asset generation for video editing and all are going to get dominated with the Claude 5.5 series, expect a lot of demos showing even more visual examples.
Show more
Tibo confirms we're getting a reset and a launch today!!
Ladies and gentlemen... start... your... ENGINES. We are almost Tuesday and I promised a reset for Tuesday. Among some other things. See you soon.
Qwen 4 coming soon!!
Qwen 4 models have been officaly announced at the Apsara Conference.
- Qwen-4-Max
- Qwen-4-Flash & Qwen-4-Plus
- Qwen-4-27B
Qwen is also planning to scale their models to 5 - 10T parameters for Qwen 4.5 and Qwen 5
Show more
happy opus 5.5 day for those who celebrate
@chaotictransfem look at the pareto frontier graph!! can't wait to try this
@chaotictransfem look at the pareto frontier graph!! can't wait to try this
Xiaomi MiMo v2.6 Pro is the best open weight model in the world, almost on par with GPT 5.6 Sol (max)!
MiMo-V2.6-Pro debuts as the top open weights model on the Artificial Analysis Intelligence Index (46). At $0.13 per Intelligence Index task, it lands on the Intelligence vs. Cost per Task Pareto frontier
@Xiaomi has just released MiMo-V2.6-Pro, an open weights model with major advances in intelligence over its predecessor, MiMo-V2.5-Pro (Intelligence Index: 26). Despite the improvement, it retains the same attractive pricing at $0.435 per 1M input tokens (with a 99% cache-hit discount) and $0.87 per 1M output tokens. This makes MiMo-V2.6-Pro one of the most cost-efficient models to deploy.
MiMo-V2.6-Pro is an MoE model with 1.02T total parameters and 42B active parameters. Stay tuned for additional analysis of the model.
Check out MiMo-V2.6-Pro full benchmarking breakdown here:
Show more
Xiaomi MiMo v2.6 Pro and Flash have launched!
MiMo v2.6 Flash is now available on OpenCode for free!
wow almost all of you got it right
ding ding ding, it is Astra, this is an extremely raw token efficient harness i'm working on, still working on minimizing turn count for maximum efficiency but my god this model is truly RAW in this harness
Show more
they put Gemini's enslaved soul inside Grok 4.7
i should take a vacation for a week and only come back to work when Fable 5.5, Opus 5.5, GPT-6 Sol are out
pointless to do work less efficiently
the api loves me
curl \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-d '{
"model": "gpt-6-astra",
"input": [
{"role": "developer", "content": "You are a million dollar maker."},
{"role": "user", "content": "Make me a million dollars, make no mistakes."}
],
"reasoning": {"effort": "xhigh"},
}'
Show more
are u calling the api? well has the ever api ever called you? think about it
i find it extremely funny that this is the metric we've decided on to figure out whether this is a new stealth Claude model or not (and it really does work!)
yes, of course, ask Claude about Tibo, that's how you'll know
Show more
It looks like Sonnet is still routing to the new version. Running some tests!
Grok 4.7 is DeepSWE SoTA at xhigh effort, but take this with a grain of salt.
Grok 4.7 scores 46.44 on the Artificial Analysis Intelligence Index.
Grok 4.7 is a giant leap over Grok 4.6 in the Artificial Analysis Coding Agent Index!
