Pinned Tweet
C++ port of TypeScript (targeting TS6) is now 100% at parity.
It's 25-750x faster and 5-20x lower RSS.
Matches TS6 1:1 on every test, every diagnostic is identical given the same inputs, and it is identical on all of the dozens of popular TS projects we compare on.
I'm like 80% sure I know exactly what Mirendil is doing here
We ran @mirendil's new base model on Baseten's private codebases; that is, code it's never seen. Mirendil-Base-Large had the lowest bits per byte (a tokenizer-independent measure of perplexity) of any model we tested.
We're excited to be the inference partner to serve Mirendil models. Congrats to the Mirendil team on this launch!
Okay I'm surprised, I ran an analysis of @pierrecomputer's new Highlights library against our in-house code editor's C++/WASM-based highlighting engine and Highlights is not faster nor deeper across the board... actually it's quite close to a tie and we're a little faster. (1/2)
So I think probably I'm gonna keep going with our own.
Where Highlights does win: less bundled bytes per language supported, by a little bit, and slightly closer agreement to Shiki for some less common languages.
Also, Highlights is open source and available now, so use it!
It's October 23 2026 and Facebook and a three-year-old Chinese startup founded by a guy who was not particularly rich in October 2023 are both ahead of OpenAI, Google, SpaceX, Alibaba, Microsoft, and Mistral on AAII
Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55, making large gains in agentic performance; it will be the leading open-weights model once its weights are published
@Kimi_Moonshot's new reasoning model gains 11 points over Kimi K3 (max, 44). It trails only Claude Opus 5.5 (max with fallback, 58), Claude Sonnet 5.5 (max with fallback, 56), and Muse Wonder Full (max, 56) on the Intelligence Index.
Once the weights for K3.1 Pro are released it will be the highest-scoring open weights model and the gap between the leading open weights and proprietary models on the Intelligence Index will narrow from 12 points to 3.
Key results:
➤ Large gains in agentic performance: Kimi K3.1 Pro scores 3rd on GDPval-AA with 1,846 Elo, up 312 Elo from Kimi K3, behind only Claude Opus 5.5 (1,866 Elo) and Muse Wonder Full (1,857 Elo). Kimi K3.1 Pro also achieves the highest score of any model on Terminal-Bench 4.0. K3.1 Pro scores 67%, up from 13% for Kimi K3 and ahead of Claude Sonnet 5.5 (64%). AutomationBench-AA, which tests agentic SaaS workflows, rises from 58% to 73%.
➤ Gains in other evals are more moderate: Humanity's Last Exam rises from 47% to 52%, CritPt from 23% to 25%, and SciCode from 59% to 61%. AA-Omniscience Index rises from 20 to 30, as the hallucination rate falls from 53% to 44% and accuracy rises from 48% to 51%.
➤ Cost per Task roughly equal to Kimi K3 at $1.98 per Task: K3.1 Pro uses 55K output tokens per Intelligence Index task, 14% more than Kimi K3 at 48K, while its first-party token prices are 33% lower.
Additional model details:
➤ Context window: 1M tokens.
➤ Size: 2.8T total, 104B active parameters
➤ Input modalities: Text and image
➤ Weights and license: Weights currently not published, Moonshot is expected to release the weights soon after launch
➤ Pricing: $2/$10/$0.20 per 1M input/output/cache hit tokens, 33% lower than Kimi K3 on all three rates.
Congratulations to the team at @Kimi_Moonshot on the release!
Beats Opus 5.5 at a lot of things, beats GPT-6 Astra at a lot of things... woah.
Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model.
🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal.
🔹 Full-project coding: games in Unreal, Unity & Godot.
🔹 Excels at parametric 3D modeling in Blender & CAD.
🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%.
🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks.
Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API.
Open weights by Nov 4, 2026.
🔗 API: platform.kimi.ai
🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro
Running a full build of ts-morph with standard settings:
ts6 - 246 seconds
rts - 1 second
I'm gonna take your CI/CD from minutes to a blink!
I understand the frustration with no early / WIP open source for this: but understand me, I didn't start this project to open source it. I'm building it for my own use first.
I think open sourcing once it's polished and good to go is an obvious win/win so that's what I intend.
We've all been bamboozled in a shocking 16-year+ plot by @rauchg
Jam stack, React, AI... was it all just a ploy? So that we'd arrive here?
WebSocket support in Vercel Functions is now generally available.
Vercel now serves 15 million WebSocket connections a day across AI chat, streaming, and realtime multiplayer apps.
vercel.com/changelog/websock…
Weird: I'm seeing bun check (the most recent one!) actually 15% *slower* than tsc 7.1 beta on my machine
This is the second time this year LiteLLM has compromised its millions of user credentials this year alone. They had a similar incident this March!!!
Please: STOP using open source software with gigantic supply chains of transitive dependencies via pip, npm, cargo, etc.
Kudos to companies like @vercel @tailwindcss etc for keeping your packages on minimal or zero dependencies.
This is more critical now than ever.
I'll write about this in a more formal setting soon!