Releases·Model Releasefeatured
Anthropic releases Claude Fable 5.1 and Mythos 5.1 with cheaper cache reads and lighter safeguards
Anthropic released Claude Fable 5.1 and Mythos 5.1 on September 1, 2026, positioning the update around lower cost for agentic work, longer task autonomy, and reduced over-triggering of safeguards. Fable 5.1 keeps Fable 5 pricing while cutting usage-billing cache reads to $0.25/MTok from $1/MTok. Alongside the release, Anthropic announced Trusted Access programs for Mythos 5.1 — a Cyber Verification Program and Life Sciences Verification Program — that reduce safeguards for defensive security work.
32079 engagement11 sourcesTue, Sep 1, 2026, 07:39 PM
Developments
Anthropic's Trusted Access programs for Mythos 5.1 reduce safeguards for defensive security work, announced with the Sept 4 release page.
Fri, Sep 4, 2026, 03:16 AM
Trusted Access programs for Mythos 5.1 announced, reducing safeguards for defensive security work including pentesting and bug bounty.
Fri, Sep 4, 2026, 03:16 AM
AI Overview
- Cache reads drop from $1/MTok on Fable 5 to $0.25/
- Anthropic reset 5-hour and weekly usage limits for all users alongside the launch.
- The company says Fable 5.1 gets further into long tasks before needing input, signals better when stuck, and writes more naturally.
- Cyber safeguards were improved so Fable 5.1 can scan users' own source code for vulnerabilities; cyber-related fallbacks to Opus are down about 40% from Fable 5 and 55% from Fable 5's launch.
- Mythos 5.1 remains available only to registered Anthropic partners in cybersecurity or life sciences; Fable 5.1 is available via API and cloud platforms.
- Anthropic says the upgrade manages costs while improving safety measures to keep the model 'safely accessible'.
- Anthropic now offers zero data retention, letting clients run Fable 5.1 on their own infrastructure; Enterprise Frontier Safeguards reach general users in the fall with client-controlled monitoring.
- The models set records on Terminal-Bench 4.0 and Humanity's Last Exam, and Anthropic released three pre-release scientific findings, including a GPU optimization and a high-resolution Venus map.
Entities
Related
Social·Communityfeatured
Opus 5.5 in the release notes
Users share first impressions of Anthropic's Opus 5.5, released September 22, praising its speed, low token consumption, and improved tone versus recent Opus models. Independent benchmarks report 100% pass rates on 23 graded tasks at roughly a third of Fable 5.1's estimated cost, while a launch-numbers roundup cites Anthropic claims of ~40% lower running costs than Opus 5 and Terminal-Bench 4.0 at 66.4%.
Users share first impressions of Anthropic's Opus 5.5, released September 22, praising its speed, low token consumption, and improved tone versus recent Opus models. Independent benchmarks report 100% pass rates on 23 graded tasks at roughly a third of Fable 5.1's estimated cost, while a launch-numbers roundup cites Anthropic claims of ~40% lower running costs than Opus 5 and Terminal-Bench 4.0 at 66.4%.
14116.3k engagement18 sourcesredditTue, Sep 22, 2026, 04:17 PM
Developments·Benchmark
Benchmarks for Anthropic's Fable 5.1 shared by Felix Rieseberg show incremental gains
Felix Rieseberg of Anthropic published further benchmark results for Fable 5.1 on X/Twitter. Commenters characterize the results as an incremental step forward rather than a leap. Some expect OpenAI's Astra to surpass these numbers soon.
Felix Rieseberg of Anthropic published further benchmark results for Fable 5.1 on X/Twitter. Commenters characterize the results as an incremental step forward rather than a leap. Some expect OpenAI's Astra to surpass these numbers soon.
760 engagement2 sourcesreddit, xTue, Sep 1, 2026, 06:50 PM
Releases·Model Releasefeatured
Anthropic releases Claude Fable 5.1 across AWS Bedrock, GitHub Copilot, and Claude Platform
Anthropic announced Claude Fable 5.1 on September 1, 2026, positioning it as frontier intelligence for coding, scientific research, and enterprise workflows. The model is available via Claude Code, the Claude Platform (claude-fable-5-1), Amazon Bedrock, and GitHub Copilot. Anthropic designated it a Covered Model with added safety and data-retention policies, and published a prompting guide plus an updated claude-api skill for migration.
Anthropic announced Claude Fable 5.1 on September 1, 2026, positioning it as frontier intelligence for coding, scientific research, and enterprise workflows. The model is available via Claude Code, the Claude Platform (claude-fable-5-1), Amazon Bedrock, and GitHub Copilot. Anthropic designated it a Covered Model with added safety and data-retention policies, and published a prompting guide plus an updated claude-api skill for migration.
Updated Wed, Sep 2, 2026, 04:56 AM — Claude Fable 5.1 became generally available in GitHub Copilot on September 2, 2026.
0302.3k engagement4 sourcesx, reddit, rssTue, Sep 1, 2026, 06:13 PM
Social·Community
Reddit user finds no reason to stay on Claude Code over Codex, citing UI flaws and model lineup
A long-time Claude Code user who switched to OpenAI Codex for personal projects argues Codex is more polished and better priced, while finding Claude Code's VS Code extension cluttered and Opus 5 disappointing. The post asks the community whether any reason remains to prefer Claude Code, beyond possible data-privacy trust in Anthropic.
A long-time Claude Code user who switched to OpenAI Codex for personal projects argues Codex is more polished and better priced, while finding Claude Code's VS Code extension cluttered and Opus 5 disappointing. The post asks the community whether any reason remains to prefer Claude Code, beyond possible data-privacy trust in Anthropic.
212 engagement1 sourceredditTue, Sep 15, 2026, 11:26 PM
Social·Communityfeatured
The new usage limits make subscription and team plans genuinely useless for real work
Claude Max users report the shift from the +50% weekly-limit promo to a permanent +25% on Sep 14 feels far harsher than Anthropic's announced ~17% cut. Data-driven analyses attribute much of the perceived loss to Fable 5.1 consuming ~31% more tokens per prompt, subscription plans not getting Fable 5.1's cache-read price discount, and an open subagent cache bug. A separate 65-day analysis alleges Anthropic quietly slashed Fable 5's thinking budget, with 39% of Claude Code calls receiving zero thinking tokens by August.
Claude Max users report the shift from the +50% weekly-limit promo to a permanent +25% on Sep 14 feels far harsher than Anthropic's announced ~17% cut. Data-driven analyses attribute much of the perceived loss to Fable 5.1 consuming ~31% more tokens per prompt, subscription plans not getting Fable 5.1's cache-read price discount, and an open subagent cache bug. A separate 65-day analysis alleges Anthropic quietly slashed Fable 5's thinking budget, with 39% of Claude Code calls receiving zero thinking tokens by August.
Updated Mon, Sep 28, 2026, 12:26 PM — A 65-day analysis of 43,000+ Claude Code invocations alleges Fable 5's thinking budget was silently cut 18-50% in August
0093.5k engagement9 sourcesredditWed, Sep 16, 2026, 12:37 AM