Marktechpost

Anthropic has released Claude Fable 5.1 and Claude Mythos 5.1, three months after the Fable 5 line shipped in June 2026. The two are the same underlying model behind different safeguard layers. Fable 5.1 is generally available as claude-fable-5-1; Mythos 5.1 stays restricted to vetted organizations. Both carry a 1M token context window and 128K max output tokens, with adaptive thinking always on.

Quantitative research agents that write their own experiments can corrupt the evidence they later learn from. A leaky feature that scores well gets stored as a successful precedent and propagated through later iterations. Prompt-level instructions and reviewer agents do not close this, because author and reviewer share the same blind spots. A team of researchers from Princeton University, Ant Group and Stanford University propose AQuA.

Voice agents fail on exactly the parts of a call that matter most: the order number, the callback digits, the email address the caller has to write down. Gradium AI has released a new text-to-speech model and made it the default across its API and Studio. The company reports an 81.0% human-rated pass rate on a 500-sentence hard-case set spanning five languages, ahead of Cartesia Sonic 3.6 at 75.1% and ElevenLabs v3 Conversational at 65.4%.

How do you benchmark a web search API when the thing being tested can read the answer key? A search agent has a fetch tool. If the gold labels sit in a public dataset, the agent can download them mid-evaluation and skip retrieval entirely. A similar problem arises when the answers are already encoded in the model’s parametric memory: a correct response no longer demonstrates that web search worked.

Google Research has released TimesFM-3, a 330 million parameter time series foundation model that forecasts multiple related series in a single forward pass. Every TimesFM checkpoint through 2.5 was univariate: one series, its own history, nothing else. TimesFM-3 is pretrained natively for multivariate forecasting on more than 1 trillion time points, and accepts multiple targets, past covariates, and past-future covariates with no task-specific fine-tuning.

The OpenClaw team has just released OpenClaw 2.0. The team shipped nothing for nearly seven weeks, after 106 releases in the previous 230 days. The update rewrites installation and rebuilds the browser Control UI. It also moves sessions and transcripts into SQLite and adds shared cloud sessions. Yes for single-operator and single-team deployments. No for multi-tenant products. Guided setup now looks for AI access already on the machine.

Time to first token (TTFT) is the metric teams use to pick an inference API for voice. It is also the metric that misleads them. TTFT marks when generation starts; a text-to-speech model cannot speak until a full clause arrives. Between those two points sits the difference between an agent that feels conversational and one that gets interrupted. This piece benchmarks every layer of the voice stack including LLM, speech-to-text, text-to-speech, and speech-to-speech.

A team of researchers from Google Cloud AI Research, Washington University in St. Louis and UNC Chapel Hill has released EnvHarness, a programmable layer that turns a static agent benchmark into one that adapts to the policy training on it. LLM agents now learn less from curated text and more from interactive environments, but those environments are hand-built and frozen: they behave identically no matter which agent is acting or how much it has improved.

Anthropic has opened a research preview of the Model Hardware Standard (MHS), a shared specification that lets AI agents discover and operate physical devices. The problem it targets is plumbing. A lab bench or factory cell is assembled from vendors that never planned to interoperate, so every instrument ships its own interface and specialists hand-write bespoke translators between each pair.

MirroS released Code-as-World: a paradigm that represents physical worlds through executable world representations. The argument is narrow and testable: pixels are evidence of a physical scene, not its ontology. A video model can predict plausible frames without ever representing mass, contact, or gravity.