%PDF-1.4 %âãÏÓ 1 0 obj << /Type /Catalog /Pages 2 0 R >> endobj 2 0 obj << /Type /Pages /Count 2 /Kids [5 0 R 7 0 R] >> endobj 3 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >> endobj 4 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica-Bold >> endobj 5 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 6 0 R >> endobj 6 0 obj << /Length 5389 >> stream BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 789.89 Tm (Anthropic Launches Claude Opus 5.5 With Fable) Tj ET BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 762.89 Tm (Level Performance at Lower Cost) Tj ET BT /F2 11 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 725.89 Tm (TechRounder PDF Edition) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 709.89 Tm (Live article:) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 697.39 Tm (https://www.techrounder.com/ai/anthropic-launches-claude-opus-5-5-with-fable-level-performance-at-lower-cost/) Tj ET q 0.82 0.85 0.9 RG 1 w 46 678.89 m 549.28 678.89 l S Q BT /F1 10 Tf 0.24 0.27 0.32 rg 1 0 0 1 46 666.89 Tm (By Vipin PG | Published September 23, 2026 | Updated September 23, 2026 | Format: Analysis | 3 min read) Tj ET BT /F2 13 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 643.89 Tm (In brief) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 623.89 Tm (Anthropic has launched Claude Opus 5.5, the first model in its Claude 5.5 family. The company says the) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 608.89 Tm (model performs at the level of Claude Fable 5.1 on most work, while costing 40% less to run than) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 593.89 Tm (Opus 5.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 568.89 Tm (Anthropic has launched Claude Opus 5.5, the first model in its Claude 5.5 family. The company says the) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 553.89 Tm (model performs at the level of Claude Fable 5.1 on most work, while costing 40% less to run than) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 538.89 Tm (Opus 5.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 516.89 Tm (The release arrived on September 22, 2026, after Anthropic CEO Dario Amodei called for the pace of) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 501.89 Tm (frontier AI development to slow enough for safety work to keep up. Opus 5.5 shows what that) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 486.89 Tm (approach looks like in practice: a more capable model with lower serving costs, faster output and) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 471.89 Tm (tighter controls around high-risk work.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 443.89 Tm (Claude Opus 5.5 targets long, demanding work) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 419.89 Tm (Anthropic positions Opus 5.5 as its strongest model so far for agentic coding, computer use and) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 404.89 Tm (professional knowledge work. The company says external testers used it for tasks such as migrating) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 389.89 Tm (a 680,000-line codebase and auditing a 200,000-line codebase in a few hours.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 367.89 Tm (Those examples come from Anthropic's own testing and customer evaluations, so they should be) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 352.89 Tm (treated as reported results rather than guarantees for every project. The broader point is easier to) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 337.89 Tm (verify: Opus 5.5 is designed for work that runs across many steps, uses tools and needs the model to) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 322.89 Tm (retain useful context over a long session.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 300.89 Tm (Anthropic's published results include a 66.4% score on Terminal-Bench 4.0 for agentic coding, an Elo) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 285.89 Tm (score of 1846 on GDPval-AA v2.1 for professional work and an 81.8% partial score on OSWorld 2.0) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 270.89 Tm (for computer use. Anthropic also reports that Opus 5.5 beats or matches several of its comparisons) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 255.89 Tm (while using fewer tokens.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 227.89 Tm (The largest change may be the price) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 203.89 Tm (Opus 5.5 costs $4 per million input tokens and $20 per million output tokens. Opus 5 was priced at $5) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 188.89 Tm (for input and $25 for output. Cache reads, which matter in coding and other agentic workloads, fall) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 173.89 Tm (from $0.50 to $0.20 per million tokens.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 151.89 Tm (Anthropic says its typical workloads cost 40% less with Opus 5.5. The company also says the model) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 136.89 Tm (produces output more than 30% faster than Opus 5. Lower token rates do not automatically make) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 121.89 Tm (every application cheaper, since the final bill also depends on context size, retries, tool calls and how) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 106.89 Tm (much work an agent completes. Even so, the combination gives developers more room to run longer) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 91.89 Tm (jobs before cost becomes the main constraint.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 1 of 2) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/anthropic-launches-claude-opus-5-5-with-fable-level-performance-at-lower-cost.pdf) Tj ET endstream endobj 7 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 8 0 R >> endobj 8 0 obj << /Length 4676 >> stream BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 789.89 Tm (Anthropic is also increasing five-hour usage limits for Pro, Max, Team and seat-based Enterprise) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 774.89 Tm (plans. Subscription users receive a rate-limit reset that can be saved and used when needed.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 746.89 Tm (Safety controls stay tied to capability) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 722.89 Tm (Anthropic says Opus 5.5 achieved its best results so far on the company's automated behavioral audit,) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 707.89 Tm (which tests the model across thousands of simulated scenarios. The company also reports stronger) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 692.89 Tm (resistance to prompt injection and says the model attempted to cross containment boundaries around) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 677.89 Tm (85% less often than Opus 5 or Claude Mythos 5.1 in a new evaluation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 655.89 Tm (Anthropic has not presented those results as proof that the model is safe in every environment. The) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 640.89 Tm (company says Opus 5.5 can sometimes recognize that it is being evaluated, which makes it harder to) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 625.89 Tm (predict how it will behave across the full range of real deployments.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 603.89 Tm (Because Opus 5.5 is comparable to Mythos 5.1 in biology and cybersecurity, Anthropic is releasing it) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 588.89 Tm (with safeguards similar to those used for Fable 5.1. Many cybersecurity tasks are routed to Opus 4.8.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 573.89 Tm (Anthropic says it will expand access through its Cyber Verification Program in the coming weeks,) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 558.89 Tm (allowing verified cybersecurity practitioners to use Opus 5.5. Biology research access is being) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 543.89 Tm (handled through the company's Life Sciences Verification Program.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 515.89 Tm (Benchmark scores need the surrounding system) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 491.89 Tm (Opus 5.5's results are also a reminder that modern AI performance depends on more than the model) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 476.89 Tm (name. Anthropic's tests use settings such as adaptive thinking, tool access, production safeguards and,) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 461.89 Tm (in some high-risk areas, fallback models. Those choices affect both the score and the cost.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 439.89 Tm (Anthropic says Opus 5.5's benchmark margins are becoming a less reliable guide to the difference) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 424.89 Tm (users will feel in daily work. That is why the cost-per-task claim matters alongside the leaderboard) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 409.89 Tm (results. A model that completes a coding or research job with fewer tokens and fewer retries can be) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 394.89 Tm (more useful to a team even when the benchmark gap is small.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 366.89 Tm (Availability and what comes next) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 342.89 Tm (Claude Opus 5.5 is available through Anthropic's platform and through Amazon Web Services, Google) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 327.89 Tm (Cloud and Microsoft Azure. Developers can use the model with the identifier `claude-opus-5-5`.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 312.89 Tm (Anthropic says Claude Sonnet 5.5 and Claude Haiku 5.5 will follow in the coming weeks.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 290.89 Tm (For developers, the sensible way to assess the release is to test it against one long-running job that) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 275.89 Tm (currently loses context, needs repeated retries or becomes too expensive to run. That result will say) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 260.89 Tm (more about the upgrade than a single benchmark percentage.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 2 of 2) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/anthropic-launches-claude-opus-5-5-with-fable-level-performance-at-lower-cost.pdf) Tj ET endstream endobj xref 0 9 0000000000 65535 f 0000000015 00000 n 0000000064 00000 n 0000000127 00000 n 0000000197 00000 n 0000000272 00000 n 0000000414 00000 n 0000005854 00000 n 0000005996 00000 n trailer << /Size 9 /Root 1 0 R >> startxref 10723 %%EOF