Software Engineer | LLM hobbyist | Self-hosting my own AI infra | Problem Solver
Florida
Joined October 2023
- Tweets549
- Following195
- Followers113
- Likes1.3K
Chris Scott retweeted
I can't trust a benchmark that ranks Grok, Gemini, Terra, and Muse above DeepSeek Flash, and by a mile.
Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks reflecting real-world usage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
Chris Scott retweeted
our next dev meetup will be nov 19!
register to attend or submit a 5 minute talk about anything you're stoked about in the dev world right now!
luma link hereee: luma.com/vjkq92vm
Chris Scott retweeted
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
Chris Scott retweeted
It's going to be about a lot more than recipes really soon.
Stay tuned. We're cooking hard. 👨🍳
Chris Scott retweeted
Our team created the technology behind CSF to give your local machines even more capacity for long workflows, large swarms and other tasks.
It's lossless, too. Meaning you don't lose any of the data you are caching by using it.
Enjoy your new free KV cache tokens!
Get up to 30% more KV cache using our lossless CSF version of @deepseek_ai V4.1-Flash-1017.
No drop in quality, just more room for you to do what you want to do!
Find the weights on our @huggingface page:
huggingface.co/local-inferen…
Chris Scott retweeted
you can now return custom HTTP denial messages to your agent or harness when a network or secret policy blocks a request in a microsandbox. 🛡️
tell it what to do next, or disable responses to close the connection without an explanation.
ALT typescript example configuring custom HTTP denial messages in microsandbox for blocked network access and secret use. below, a 403 forbidden json response contains the code “secret_policy_denied,” the message “review secret access.” and the domain “example.com.”
Chris Scott retweeted
FLUX 3 Image Klein is here.
Open Weights with the quality of a large frontier model. Customizable and fast.
5.6B parameters and runs on a single consumer GPU.
Native 4K, bounding boxes, and pixel-perfect editing.
Chris Scott retweeted
GLM 5.3 Flash, DeepSeek V4.1 Flash, Qwen 3.8 Next Flash, and even Qwen 3.8 27B are all outperforming (in both intelligence and capabilities) every model that was considered "frontier intelligence" in Xmas 2025 (just 10 months ago)
Opensource AI is on fire
Chris Scott retweeted
The quality of the attendees applying so far is insane.
We will sell out so get in your application early 👀
Ticket applications are OPEN for AI Engineer Code Summit SF!
Join us for 3 days of deep technical learning alongside the pioneers and innovators shaping the future of AI.
Only 1,000 applications will be accepted.
Apply today ➝ ai.engineer/code
Chris Scott retweeted
turn out when you post about your meetup more people show up, who would have known! 🤣
tomorrow, @thelabmiami we’re cohosting @OmarchyLinux Miami No. 1 with @davidfano @RalphQuintero @GianniDalerta come through and bring you old laptop, Mac mini, or raspberry Pi, we’ll help you set it up!
Chris Scott retweeted
You can now scrape the web with AI agents in 85ms.
Someone built in Rust what Chrome engineers couldn't optimize in years.
Obscura is an open-source headless browser engine for web scraping and AI automation.
A 70MB binary that boots instantly, loads pages in 85ms, and runs on 30MB of RAM.
The Google equivalent needs 300MB of disk, 200MB of memory, and 2 seconds to start.
The real trick is stealth mode.
Per-session fingerprint randomization across:
> GPU, canvas, audio, battery
> 3,520 tracker domains blocked
> navigator.webdriver fully masked
> Native functions patched cleanly
No detector can catch it.
The engine runs real JavaScript through V8 and speaks the Chrome DevTools Protocol natively.
That means it works as a drop-in for Puppeteer and Playwright.
You swap the binary, your code stays untouched.
One developer shipped this alone. Free to use.
Chris Scott retweeted
🐦Tokens now go brrrr
🪽Ornith-1.5 models just got new wings: MTP weights have been updated for the BF16, GGUF, NVFP4 & FP8 variants.
🔗huggingface.co/collections/o…
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies.
It achieves state-of-the-art performance among open-source models of comparable size and delivers performance comparable to Claude Opus 4.8 across reasoning, agentic, and coding tasks:
✅Terminal-Bench 2.1 (86.1)
✅SWE-Bench (86 on verified, 65.1 on pro, 79.6 on Multilingual)
✅DeepSWE (56)
✅HLE (44.6)
✅ClawEval (81.4)
✅Tool Decathlon (71.2)
Ornith-1.5 takes a major step toward training foundation models through end-to-end self-improvement, extending the self-scaffolding strategies introduced in Ornith-1.0 into a more complete self-improvement loop: the model proposes new tasks, generates task-specific scaffolds, and produces solution rollouts for reinforcement learning, continuously creating new learning experiences from which it can improve.
All models, along with their quantized versions (FP8, GGUF, MLX, and NVFP4), have been released under the MIT License, enabling unrestricted commercial and research use.
📘Tech Blog: ornith.ai/ornith_1_5.html
🤗Huggingface: huggingface.co/collections/o…
