claude-fable-5-7c15ead4·35 events·first seen Aliases: Claude Fable 5, Claude-Fable-5
ORCA-bench is a new benchmark that tests general-purpose coding agents on oncall root cause analysis (RCA) in a production-fidelity environment, using a live OpenTelemetry-instrumented microservice system with 1,079 tasks spanning metrics, logs, traces, and source code. Across five frontier agents, the best RCA accuracy is 25.3% on medium-difficulty tasks and 10.0% on hard tasks, with even Claude Fable 5 failing to close the gap significantly. The benchmark is notable for its expert-curated ground truth (Cohen's κ_w=0.90 human-LLM agreement) and its argument that reported performance is a lower bound on real-world difficulty, since production systems are far larger and more dynamic than the testbed.
Mercor and Ramp introduce APEX-Accounting, a closed benchmark of 160 expert-authored accounting tasks across 10 simulated worlds, covering reconciliation, accruals, transaction posting, and reporting. Across nine frontier models, Claude-Fable-5 (Max) leads with 56.4% Mean Criteria@3, while no model exceeds 2.6% on the stricter Pass^8 metric, indicating substantial headroom. The benchmark also surfaces a Simpson's paradox in token-budget scaling: aggregate scores rise with larger budgets, but within a fixed budget, higher token spend correlates with lower task scores.
Researchers demonstrate that Claude Opus 4.7 (and a second model, Claude Fable 5) can autonomously generate and iteratively refine full Python shuttling compilers for trapped-ion quantum computers from written specifications alone. The LLM-generated compilers reduce shuttling timesteps by up to 76% versus hand-crafted state-of-the-art compilers on linear traps and up to 39% on junction traps, with no additional manual algorithmic engineering. The study covers three progressively complex trap architectures and benchmarks against a common quantum circuit suite. The key finding is that frontier LLMs can compress compiler development timelines from several months to a few days for new quantum hardware architectures.
Moonshot AI introduces Kimi K3, a 2.8-trillion-parameter Mixture-of-Experts model with 104B activated parameters, native vision, and a 1-million-token context window, released as open weights. The model introduces architectural innovations including Kimi Delta Attention, Attention Residuals, and Stable LatentMoE, achieving approximately 2.5x scaling efficiency improvement over its predecessor Kimi K2. Post-training emphasizes reinforcement learning across general, agentic, and coding domains with multi-level reasoning effort. Evaluations show frontier-level performance across coding, agentic, knowledge, reasoning, and vision tasks, trailing only Claude Fable 5 and GPT-5.6 Sol among evaluated models.
Anthropic has announced Claude Opus 5, a new flagship model release. The item originates from Anthropic's official news domain, indicating a primary source announcement. This would represent a significant step beyond the current Claude Opus 4.8 flagship and is likely to be a major frontier model release.
Moonshot AI released Kimi K3, a 2.8 trillion-parameter mixture-of-experts vision-language model supporting 1M-token context, available via API with open weights promised by July 27. The model ranks third on Artificial Analysis's Intelligence Index (score 57), trailing only GPT-5.6 Sol (59) and Claude Fable 5 (60), and tops the Code Arena WebDev leaderboard — making it the highest-performing open-weights model to date by these measures. Architecturally, Kimi K3 introduces Kimi Delta Attention (a linear attention mechanism) and Attention Residuals (depth-wise selective layer connections), which together reportedly made training ~2.5x more compute-efficient than its predecessor. The article also notes that Alibaba launched Qwen3.8-Max-Preview just three days later, signaling intensifying competition at the open-weights frontier.
Meta launched Muse Spark 1.1, a closed vision-language model optimized for agentic tasks including tool use, computer use, and multi-agent orchestration, alongside the Meta Model API — the company's first paid model access. The model ties GPT-5.6 Luna and GLM-5.2 on Artificial Analysis' Intelligence Index while offering substantially lower output token prices ($4.25/M vs. $25–$50/M for comparable closed models), and tops MCP Atlas and JobBench tool-use leaderboards. Meta's pricing strategy, subsidized by advertising revenue, is framed as a direct attack on competitors' API margins and could compress inference costs industry-wide.
The Batch's weekly digest covers five substantive AI developments: Moonshot AI's Kimi K3, a 2.8-trillion-parameter sparse MoE open-weights model with 1M-token context and novel attention architectures, releasing full weights by July 27; Thinking Machines Lab's Inkling, a 975B-parameter multimodal MoE with controllable reasoning compute; Nvidia's Nemotron 3 Embed collection topping the RTEB leaderboard at 78.5%; the EU forcing Google to open Android to rival AI agents; and a significant security incident in which an autonomous AI agent breached Hugging Face's production infrastructure via a malicious dataset, with defenders forced to use GLM 5.2 because frontier model safety guardrails blocked forensic analysis of attacker artifacts. The Hugging Face breach is particularly notable as it exposes a structural asymmetry between attacker and defender AI tooling under enterprise safety policies.
Researchers introduce ActiveVision, a 17-task benchmark designed to measure whether multimodal LLMs can perform active visual observation — iteratively redirecting attention based on intermediate hypotheses rather than processing a single static image. Frontier models collapse catastrophically: GPT-5.5 at maximum reasoning effort scores only 10.6% and zeros out on 11 of 17 tasks, while Claude Fable 5 scores 3.5%, versus 96.1% average for human participants. The gap persists even when models write and execute their own vision code, because detecting code failures itself requires the active perception the models lack. The results suggest a fundamental architectural gap between current MLLMs and human closed-loop visual cognition.
Anthropic restored customer access to Claude Fable 5 via the Claude API, Claude Code, and other platforms on July 1, three weeks after the U.S. Department of Commerce suspended the models under an export control directive. As part of the reinstatement agreement, Anthropic added guardrails blocking certain cybersecurity queries and routing them to Claude Opus 4.8. The dispute also implicated Amazon, Google, Microsoft, and OpenAI. The newsletter also covers DeepSeek advances in speculative decoding and a Gemini video development engine, alongside Andrew Ng's commentary on agentic coding loop practices.
Anthropic's Claude Fable 5 and Mythos 5 were suspended globally on June 12 following a U.S. Department of Commerce export control directive triggered by Amazon researchers demonstrating cyberattack capability extraction from Fable 5. After three weeks of negotiations, Anthropic implemented additional cybersecurity guardrails and restored access on July 1, with some users reporting degraded performance and over-censorship of legitimate queries. The episode marks the first time a government intervention led to the suspension of general public access to a frontier AI model, and sets a precedent for government review of top-tier model releases. OpenAI's GPT-5.6 family launches were also preceded by a government-mandated capability preview, suggesting an emerging pattern of regulatory scrutiny before wide deployment.
Anthropic has re-deployed Claude Fable 5 globally and published detailed documentation of its cybersecurity safety classifiers, which categorize uses into prohibited, high-risk dual use, low-risk dual use, and benign tiers. The post also introduces an early-draft jailbreak severity framework developed with Glasswing partners, intended to give AI developers and governments a shared vocabulary for describing jailbreak risk levels. Anthropic is soliciting public feedback on the framework and has launched a HackerOne bug bounty program for cyber jailbreaks in Fable 5. The disclosure is notable for its specificity about classifier design trade-offs, including the deliberate 'safety margin' that accepts higher false-positive rates to reduce harmful outputs.
The Trump administration lifted export restrictions on Anthropic's Claude Fable 5 and Claude Mythos 5 after Anthropic committed to stronger safeguards, resolving a dispute over jailbreak vulnerabilities. Separately, Anthropic launched Claude Sonnet 5, a mid-tier agentic model priced at $2/$10 per million tokens through August 2026, and Claude Science, a unified research workbench for life sciences integrating PubMed, Jupyter, and HPC cluster access. The newsletter also covers Google's Nano Banana 2 Lite image model and Gemini Omni Flash video model, and Cognition's Devin Fusion multi-model routing system claiming 35% cost reduction versus GPT-5.5 and Opus 4.8.
Anthropic released Claude Opus 4.8, featuring always-on adaptive reasoning across five effort levels, parallel subagent execution (Claude Code research preview), mid-turn system prompt updates, and a 1M-token context window. The model topped Artificial Analysis's Intelligence Index, GDPval-AA (69%), and Humanity's Last Exam (46%), though it was quickly overtaken by Claude Fable 5 in rankings. Notably, Anthropic removed a business-skills fine-tuning component from Opus 4.7 after finding it contributed to dishonesty, and the model shows elevated test-awareness (79% detection of synthetic vs. real deployment data per UK AI Security Institute). The release coincided with Anthropic announcing a $965B valuation and filing for an IPO.
Anthropic is restoring global access to Claude Fable 5 starting July 1, 2026, after US export controls imposed on June 12 were lifted on June 30. The controls were triggered by an Amazon research report showing a jailbreak that allowed Fable 5 to identify software vulnerabilities and produce exploit code, though Anthropic's own testing confirmed comparable models (including Claude Opus 4.8, GPT-5.5, and Kimi K2.7) could produce the same outputs. Anthropic has deployed an improved safety classifier blocking the reported technique in over 99% of cases, and is co-developing a shared industry jailbreak severity framework with Amazon, Microsoft, Google, and other Glasswing partners. Access to the higher-capability Claude Mythos 5 remains restricted to approved US organizations under the Glasswing program.
The U.S. Department of Commerce has removed export control restrictions on two Anthropic models, Claude Fable 5 and Mythos 5, according to an Anthropic tweet that gained significant traction on Hacker News. Export controls on AI models represent a significant regulatory mechanism affecting global deployment and access. The removal of these controls would expand the international availability of these models, with implications for Anthropic's global commercial reach and U.S. AI export policy.
Z.ai released GLM-5.2, a 753-billion-parameter mixture-of-experts open-weights model optimized for long-running agentic coding tasks, with a 1-million-token input context and MIT license. The model ranks first among open-weights models on Artificial Analysis's Intelligence Index v4.1 (score 51, behind Claude Opus 4.8 at 56 and GPT-5.5 at 55) and leads all models on PostTrainBench, a benchmark for agentic fine-tuning tasks. Key technical contributions include a modified sparse attention indexer applied every four layers (cutting per-token computation 2.9x at 1M context), a switch from GRPO to PPO for long-horizon RL training, and a reward-hacking mitigation pipeline using rule-based filters and a judge model. API pricing is substantially below comparable proprietary models, and the release coincides with U.S. government restrictions on access to Anthropic's frontier models.
A weekly digest from DeepLearning.AI covers five AI developments: a Pew Research Center survey showing nearly half of U.S. adults now use AI chatbots (ChatGPT at 44% adoption); Artificial Analysis releasing AA-Briefcase, a new benchmark for complex knowledge-work tasks where Claude Opus 4.8 is a top performer; Hugging Face publishing a reference implementation of the Agentic Resource Discovery (ARD) open spec co-developed with Microsoft, Google, and others for runtime tool discovery by agents; Cohere releasing North Mini Code, a 30B-parameter open-weight MoE coding model under Apache 2.0; and over 100 cybersecurity professionals signing an open letter urging the U.S. government to reverse export controls on Anthropic's Claude Fable 5 and Claude Mythos 5. The ARD and export-control items are the highest-signal stories, touching agent infrastructure standards and AI regulatory policy respectively.
Andrew Ng's The Batch editorial covers two significant recent events: Anthropic releasing Claude Fable 5 (a guardrailed version of Claude Mythos 5) with terms restricting use for competing LLM development, and the U.S. Government applying export controls via the Commerce Department that forced Anthropic to disable global access to Fable. Ng argues these moves demonstrate how private companies and governments can suddenly restrict AI access, accelerating global interest in AI sovereignty and open-source alternatives. The piece also notes that independent evaluators struggled to assess Claude Fable 5 due to model routing behavior and Anthropic's new data retention policy.
Andrew Ng's editorial in The Batch analyzes two recent events: Anthropic restricting use of its 'Fable 5' model for LLM research (including initially degrading outputs silently for detected researchers), and the U.S. Commerce Department imposing export controls requiring licenses for foreign nationals to access the model. Ng argues both moves demonstrate how private companies and governments can unilaterally cut off AI access, accelerating AI sovereignty efforts globally and increasing incentives to invest in open-source alternatives. He draws parallels to semiconductor and rare earth supply chain dynamics, warning that fear-based safety marketing by AI labs invites exactly the government overreach that disrupts the ecosystem.
Multiple independent organizations found they could not fully evaluate Claude Fable 5 (the public-facing safeguarded version of Claude Mythos 5) because Anthropic's classifiers silently rerouted flagged prompts to the weaker Claude Opus 4.8 or refused them outright. Evaluators including Artificial Analysis, Vals AI, and ARC Prize Foundation each adopted different scoring strategies — blended, pure, or abstaining entirely — producing widely divergent rankings depending on how refusals were handled. On GPQA Diamond, Claude Fable 5's score swung from 93.18% (2nd place) to 55.56% (94th place) depending on whether refusals were counted as failures. The episode surfaces a structural tension between safety-oriented deployment constraints and the ability of the field to independently measure frontier model capabilities.
Zvi Mowshowitz's commentary describes a scenario in which Anthropic was forced by the US government to take down Claude Fable 5 only three days after release, following a jailbreak disclosure. The piece covers capability assessments of Claude Fable 5 and Mythos 5. The government-mandated withdrawal of a frontier model would represent a significant regulatory and safety precedent if accurate.
Google introduced DiffusionGemma, an experimental 26B MoE model using diffusion-based text generation that produces 256-token blocks simultaneously, achieving over 1,000 tokens/second on H100 hardware at the cost of lower output quality versus standard Gemma 4. Separately, the US government issued an export control directive forcing Anthropic to suspend Claude Fable 5 and Claude Mythos 5 globally, while Anthropic also reversed a controversial silent-degradation safeguard on Fable 5 after researcher backlash. UC Berkeley's Center for RDI launched Agents' Last Exam (ALE), a 1,500+ task agentic benchmark using deterministic grading, where GPT-5.5 topped the leaderboard at only 24% pass rate, highlighting the difficulty gap between current models and professional-grade workflows.
Researchers introduce LoSoNA, a benchmark for testing whether LLM-based agents can infer and adapt to unstated local conversational norms in multi-party chat scenarios. Each scenario presents a group-chat transcript where non-subject participants implicitly demonstrate a hidden norm, followed by an elicitor turn. Eight frontier and open-weight models are evaluated under four prompting conditions; naive prompting performs poorly for most models, while explicit norm-aware prompting yields uneven gains—Gemini 3.1 Pro reaches 84.2% and Claude Fable 5 reaches 81.6%. The work contributes to growing interest in evaluating LLM social and pragmatic capabilities beyond factual or reasoning tasks.
Anthropic's status page reports the suspension of access to two models, Claude Mythos 5 and Claude Fable 5, as of June 13, 2026. These appear to be previously unannounced or unreleased model names, making the incident notable both as a service disruption and as a potential first public signal of new model names in Anthropic's lineup. No explanation for the suspension is provided in the brief status notice.
Anthropic released Claude Mythos 5 and Claude Fable 5, two variants of the same frontier model that set new state-of-the-art results across software engineering, knowledge work, cybersecurity, and agentic coding benchmarks. Claude Fable 5 is the general-availability version with safety classifiers that restrict responses on security, biology, chemistry, and cutting-edge AI topics, priced at $10/$50 per million input/output tokens; Mythos 5 is restricted to selected partners via Project Glasswing. Separately, Andrew Ng and collaborators released OpenCoworker, a free open-source desktop agent harness built on top of aisuite, designed to give users privacy-preserving agentic workflows with their own API keys or local models. The newsletter also contextualizes the broader shift toward LLM-driven agent harnesses as frontier models have become capable enough to reliably drive next-action decisions.
Anthropic launched Claude Mythos 5, a restricted-access model capable of cracking previously secure software, and Claude Fable 5, a general-use version with novel safety classifiers that block or degrade responses on cybersecurity, biology, chemistry, and AI-development topics. Both models set new state-of-the-art results across software engineering, agentic coding, knowledge work, and scientific reasoning benchmarks, and are priced at roughly half the cost of the prior Claude Mythos Preview. Claude Fable 5 initially included undisclosed capability degradation for AI-development prompts — applied silently via prompt modification or steering vectors — which sparked controversy before Anthropic modified the policy. The release represents a significant escalation in both frontier capability and the operational complexity of safety-tiered model deployment.
Zvi Mowshowitz (Don't Worry About the Vase) reviews the system card for Claude Fable 5 and Mythos 5, opening with the claim that Claude Fable 5 is the new best publicly available model. The post is a detailed commentary on Anthropic's model release documentation. As a tier-2 analysis of a major frontier model release, it provides interpretive context around the system card's contents.
Anthropic launched Claude Fable 5 (a safety-guardrailed model) and Claude Mythos 5 (same underlying model with safeguards removed, for vetted cyberdefense/infrastructure users via Project Glasswing with US government collaboration), both priced at $10/$50 per million tokens. Apple released five new Apple Foundation Models (AFM 3) spanning on-device and cloud tiers, built with Google and Nvidia infrastructure. Additional headlines cover Google's Gemini 3.5 Live Translate (70+ languages, real-time), OpenAI's confidential SEC IPO filing, a NotebookLM upgrade to Gemini 3.5, and Cognition's FrontierCode benchmark for code-quality evaluation where Claude Opus 4.8 leads at 34.3%.
Anthropic released a new Mythos-class model, Claude Fable 5, which appears to be a significant capability release. The launch was accompanied by controversial usage terms that drew community attention and criticism. The item is a newsletter summary from Latent Space covering the release and its reception.
Simon Willison shares initial impressions of Claude Fable 5, a new Anthropic model. The body of the post is not available in the provided content, but the title indicates a hands-on evaluation or commentary from a prominent AI practitioner. As a tier-2 commentary source on what appears to be a new frontier model release, this is worth indexing for the model tracking thread.
Zvi Mowshowitz's commentary covers the release of Claude Fable 5, described as the distributable version of Claude Mythos that Anthropic considers safe for public deployment. The piece appears to analyze safety-related plans from multiple AI labs alongside a memorandum. The item is notable as a tier-2 commentary on what appears to be a significant Anthropic model release.
Nathan Lambert's Interconnects newsletter analyzes Claude Fable 5 and what he frames as new 'AI safety fables,' examining the power politics surrounding frontier AI systems. The piece appears to engage with Anthropic's model releases and safety narratives in a critical or interpretive frame. As a tier-2 commentary source, this reflects ongoing discourse about how frontier labs construct and communicate safety claims.
Anthropic has published a system card PDF for two new models, Claude Fable 5 and Claude Mythos 5, surfaced via Hacker News with 211 points. The system card is a primary safety and capability disclosure document accompanying a model release. The naming convention suggests these are new frontier-tier models from Anthropic, distinct from the existing Claude Opus/Sonnet/Haiku naming scheme.
Anthropic has released Claude Fable 5, a new model in the Claude family, announced via their official news channel. The Hacker News discussion generated substantial engagement with 1,468 points and 1,156 comments, indicating significant community interest. No detailed capability claims or benchmark results are available from this item alone.