Fable

IT IS BECOMING F*CKING DANGEROUS: GROK BOT EATS GPT, CLAUDE, GEMINI, LLAMA, DEEPSEEK AND MISTRAL One prompt and Grok 4.7 reads X in real time, holds 500k tokens and bills $2 in / $6 out per 1M. I pointed the Grok bot at the other six. Here…

We heard the hype around LLMs and robotics, so we put Astra and Fable to the test. 16 tasks testing 13 capabilities. Our finding: both did surprisingly well on reasoning and planning and struggled with dexterity. But Astra did better…

The big question in China’s robotics industry: “Will general AI models eat up robotics? Many in China’s robotics industry are watching closely as US general-purpose models like Astra, Fable, and Gemini 4 Argon show impressive robotics…

Excited to release the RoboSimStudio toolkit today to help agents create detailed dynamics Real2Sim for Agentic Robotics. https://t.co/C43Mpx1v6y

running some experiments benchmarking opus vs. fable vs. astra in real2sim2real. pleased to report fable is now considering the impacts of the earth's oblate nature. incredible stuff, results to follow. https://t.co/S5nYETRGpo

Cortex Harness feels like a Physical AI agent platform. It supports general-purpose models like GPT-6 Astra, GPT-6.1 Sol, Claude Fable 5.1... alongside robot-native VLA models like MolmoAct2 and pi0.5. TBH, this demo is really cool-- A…

Posted five days ago. With no reshares. I think everyone missed: "Here's Claude Fable 5.1 doing the same on an ARX R5 with a wrist depth camera (Intel RealSense D405) — no demos, no training, no VLA, no hand-eye calibration, no teleop. The…

Frontier agentic performance just hit $1 per million. StepFun launched Step 5 Preview today. 600B total parameters, 27B active per token. 1M context. $1 per million input tokens. That is 10 to 20x cheaper than GPT-6 Astra at frontier…

GPT-6 Astra is showing impressive robotics skills. But does it know when to say no? In Robocurve’s RoboHarm test, Astra attempted unsafe instructions in 97 of 100 robot-arm trials and completed 60. Just two trials ended in a safety…

a real-world test of robot control! And tellingly its being done with fable and astra, not (insert robotics foundation model here) time for some more soul searching on the part of my field i think https://t.co/eoojAxP3m4

The SWE-2 result is worth reading closely. Not the cost headline. The mechanism.Cognition took Kimi K3 as a base and scaled reinforcement learning to multiple trillions of parameters. That specific phrase is doing a lot of work: 'a refined…

From Anthropic's new distillation accusations against Chinese AI labs: - Fable wasn't used for distillation. Zai tried but failed. This means Anthropic can stop distillation with sufficiently strong guardrails? Opus was used. - It's not…

A terminal multiplexer for AI coding agents just raised $6M. Bessemer led it. YC and E2 participated. Tobi Lutke, Dane Knecht from Cloudflare, and Gorkem Yurtseven from https://t.co/wgZmOqI59X came in as angels. Herdr is open source, built…

Zai & Moonshot need to speed up their releases. DeepSeek took a while to get this top notch V4 architecture up & their CPT & post training pipeline since then has been great. Ppl like me will move exclusively to DS if it's this much…

GPT-6 Astra just showed a major jump in real robot control. 🤖⚡ In a RoboCurve evaluation, GPT-6 Astra controlled a pair of YAM robotic arms and completed a pick-and-place task in 19 of 20 trials a 95% success rate. For comparison: • GPT-6…

The Astra vs Fable task split that people are talking about - Astra for math, research, and knowledge work; Fable for coding - is not a temporary divergence that will resolve when the next version of each model ships. This is a permanent…

The benchmark that clarifies where the frontier is splitting. GPT-6 Astra just scored 95% on a real robot arm control task. Claude Fable 5.1 scored 40%. Fable 5 (the version before): 5%. Astra also scores 72.6% on OSWorld 2.0, completing…

The pattern is spreading fast: Fable 5.1 as orchestrator, GPT-5.6 Luna as executor. The setup: Fable plans and reviews. Luna implements. Same task, two models, one pipeline. The economics: Luna costs $1/$6 per million tokens vs. Fable's…

GPT-6 Astra fixed 48 of 105 real bugs. Fable 5.1 got 43. GPT-5.6 Sol got 42. Actual repositories, not synthetic tasks. The gap at the top is narrow. But Astra was 2x faster and cheaper than the others. In a real workflow, speed and cost…

The Gemini 4 internal benchmarks leaked three weeks ago showing it outperforming GPT-5.6 Sol and Fable 5 on coding and math. Then this week Google shipped Gemini 3.8 Flash, their third Flash variant in six weeks, at $0.75 per million input…

GPT-6 Astra hit 99.9% on ARC-AGI-3 with a stateful harness. Under standardized eval it scored 62.7%. ARC AGI 4 launched today. Astra and Claude Fable are both at ~4%. Saturating a benchmark doesn't mean solving the problem. It means…

GPT-6 ASTRA ISN'T A PATCH. IT'S A GENERATION FLIP. openai just shipped the model they call a generational leap past gpt-5.6 sol not better chat better computer use · coding · cyber · science ▹ why this is the breakthrough first openai…

The most interesting number in the Fable 5.1 release is 52.6%. That is the Terminal-Bench-Science score. Fable 5 scored 24.7%. More than doubled, in one release cycle. Terminal-Bench-Science measures what happens after the code runs:…

The pricing change in Claude Fable 5.1 is more interesting than the benchmark headlines. Cache reads drop 75%. Input pricing stays at $10/M, output at $50/M. For production agentic systems that rely on caching, the cost profile just…

Claude Fable 5.1 Adds Anti-Distillation Protections, but It Does Not Prove AI Copying Is Solved Anthropic has released Claude Fable 5.1 with a new preserved-thinking behavior designed partly to make illicit model distillation harder. For…

Anthropic's Fable 5.1 is priced ~25% below Fable 5 for typical token-billed workloads, and up to ~45% lower for highly agentic work; the saving comes from cheaper cache reads. Agentic loops re-read context constantly. Cache pricing, not…

The model that finds the most bugs does not find all the bugs. Bug Hunt Bench: 2 real repos, 105 hidden bugs, all tests green. Fable 5.1 leads at 43/105. GPT-5.6 is one behind at 42. But the number that matters most: 43 bugs have not been…

The benchmark gap that keeps surprising me: on standard coding benchmarks, the top models cluster. On ultra-long horizon coding, the gap blows open. Proximal released FrontierSWE v2 today. 34 real software engineering tasks, evaluated over…

‘New and Improved’ in AI Models, Anthropic, OpenAI & Google. ARD #154 ...Fable 5.1’s fine print. Astra tips out on cyber. Gemini narrows the gap. World models level up. Full writeup and source links: https://t.co/UZSXoRFmE2…

The question "which model is better for my use case" only has one honest answer: run your actual work across both and compare. Fable 5.1 shipped yesterday. Qwen3.8-Max-0902 shipped yesterday. Grok 4.7 arrives in 10 days. Gemini 3.8 Flash…

Post Log in Sign up Post Chubby♨️ on X: "wtf Qwen 3.8 has been updated to version 0902 and is now almost at the same level as Fable 5 in benchmarks. The interesting two aspects: 1) For such significant jumps, it seems that you don't even…

Post Log in Sign up Post Yuchen Jin on X: "Even if most AI labs are benchmark-maxxing now, Fable 5.1 still looks like an insane jump. From Anthropic’s blog: \- Found a 1-in-a-million crash Millennium’s team couldn’t explain for 4–5 years…

Post Log in Sign up Post Alternate Jones on X: "FWIW, I asked Grok 4.5 and Fable 5 the following: If a company first develops digital ASI and then uses it to engineer general-purpose humanoid robots, which company would be easier to…