MODEL
Claude Opus 5.5
Overview
Claude Opus 5.5 is Anthropic‘s Sept 22, 2026 Opus-tier frontier release, framed by the company as “Claude Fable 5.1-level intelligence” at 40% lower run cost than Claude Opus 5 and >30% faster output. Per-token API pricing dropped from $5/$25 to $4/$20 per Mtok (a 20% cut), while cache reads dropped to $0.20/Mtok — a 60% cut. The “40% less to run” figure is Anthropic’s own composite; the per-token comparison is 20% and the cache-read discount is where the rest lives. Agentic workloads (multi-turn, prompt-cache-heavy) capture the full 60% cache-read cut, while single-shot API calls only see the 20%. Claude Code v2.1.280 swapped the Opus default to claude-opus-5-5 the same day.
Timeline
-
2026-09-23-AI-Digest — Anthropic releases Claude Opus 5.5 on Sept 22, framed as Claude Fable 5.1-level intelligence at 40% lower run cost than Claude Opus 5 and >30% faster output. Per-token pricing drops from
$5/$25to$4/$20per Mtok (a20%cut); cache reads drop to$0.20/Mtok— a60%cut. Anthropic also lifted five-hour usage caps on Pro/Max/Team/Enterprise seats and flagged Sonnet 5.5 and Haiku 5.5 for the coming weeks. Same-day, Claude Codev2.1.280swapped the Opus default over toclaude-opus-5-5. Lands the same day as OpenAI‘s GPT-6 Sol / GPT-6 Luna release — the sub-24h dual launch with matched per-token cuts is the coordinating datapoint. -
2026-09-24-AI-Digest — Claude Opus 5.5 is the model running the 21-hour autonomous Anthropic agent-loop that surfaced a novel RNA-repeat + reverse-transcriptase enzyme system with Feng Zhang’s on-the-record “genuinely intriguing” quote. Load-bearing framing: the agent loop (not a specialised model) produced the candidate — Opus 5.5 acts here as general-purpose reasoning substrate, not a bespoke biology model. Also referenced as reasoning-tier pricing anchor in the “Tokens too cheap to meter” HN discussion at
$4/$20, where the tier-specific-not-market-wide framing is load-bearing (Luna/Flash-tier commodity vs Opus 5.5/gpt-5-high reasoning-tier). Aider polyglot top-5 rescore hasn’t rotated to include Opus 5.5 yet — measurement lag from the Sept 22 paired-launch continues. -
2026-09-25-AI-Digest — Opus 5.5 surfaces in two HN threads today. (1) “Opus 5.5 is good at explainer videos” (~
172pts ·~100cmts, launchvideo.io) — practitioner writeup showing Claude Opus 5.5 producing end-to-end explainer videos, another datapoint that frontier LLMs are crossing into multi-modal video-production workflows practitioners are actually shipping into. (2) Opus 5.5 is again referenced as the model behind the 21-hour autonomous ART-enzyme agent loop, now under Bloomberg’s replication-doubt beat — Anthropic’s own methodology write-up disclosing 10 internal reruns that failed to rediscover the ART system. Load-bearing framing to carry: the capability demonstration (long-horizon autonomous DNA search producing a plausible novel candidate) survives; the validated-new-gene-editing-platform second-order framing does not. Opus 5.5 remains general-purpose reasoning substrate, not a bespoke biology model. -
2026-09-27-AI-Digest — Simon Willison published on 2026-09-26 a short piece using Claude Opus 5.5 to author an animated HTML5-canvas pixel-art kākāpō clip and Claude Code driving Playwright to record it — a small but concrete “Claude Code as creative-tool driver” data point for the growing corpus of agentic-coding-in-the-wild examples. Load-bearing softener: one hobbyist example does not settle the “which model is best at driving Playwright” question, and Willison’s post is a keynote demo rather than a systematic comparison. Reframe worth carrying:
Claude Opus 5.5 + Claude Code + Playwright is a working end-to-end animation pipeline in Willison's hands today, notClaude has decisively won the tool-driving-model tier. Ties to Google‘s early-Gemini-4 post-training preview roadmap flagged by DeepMind CEO Koray Kavukcuoglu at The Information’s AI Agenda Live Summit last week (already-reported:2026-09-26-AI-Digest); the two threads together sketch the tool-driving-model axis — Willison’s post is the shipped example, Kavukcuoglu’s summit statement is the roadmap. Log against MOC - Agentic Coding and MOC - Developer Tools. -
2026-10-31-AI-Digest — Claude Opus 5.5 is the named migration target for the deprecation of Claude Opus 4.8 fast mode (deprecated
2026-10-30, removed2026-11-30), per the deprecation note shipped with Claude Codev2.1.316. Standard-mode Opus 4.8 continues; only the fast-mode variant is being removed, and Opus 5.5’s20%per-token cut and60%cache-read cut (vs the prior Opus 5 price) are the lower-friction fast-tier substitute. Corpus framing to carry: Opus 5.5 absorbs Opus 4.8’s fast-mode demand on the migration side of a scheduled SKU compression, continuing the role Opus 5.5 has carried since the Sept 22 launch as the lower-cost reasoning-tier substitute positioned to pull traffic from the retained-but-older Opus 4.8 price point. Log against MOC - Agentic Coding, MOC - Developer Tools.
Key Developments
-
The “40% Less to Run” Framing Decomposes as
20%Per-Token +60%Cache Reads (September 22, 2026): Anthropic’s headline composite is real but the underlying levers differ by workload. Per-token pricing moved$5/$25→$4/$20(a20%cut) and cache reads dropped to$0.20/Mtok— a60%cut. Agentic workloads (multi-turn, prompt-cache-heavy) capture the full 60% cache-read cut; single-shot API calls only see the 20%. Anyone quoting the 40% number for isolated one-shot cost comparisons is off by 2×. -
Same-Day Substrate Cutover in Claude Code v2.1.280 (September 22, 2026): Claude Code
v2.1.280swapped the Opus default over toclaude-opus-5-5on the same tag that shipped the model — same-day model-and-substrate delivery, tightening the pattern the Claude Opus 5 /v2.1.219launch already established. -
Sub-24h Paired Frontier Launch With OpenAI (September 22, 2026): OpenAI shipped GPT-6 Sol and GPT-6 Luna the same day with matched per-token cuts. Simon Willison flagged the sub-24h dual launch as unprecedented; near-simultaneous frontier releases have clustered before, but a sub-day dual launch with paired price cuts hasn’t. The coordinating framing to carry:
coordinating datapoint on the price curve, notprice war ignition. -
Five-Hour Usage Caps Lifted on Pro/Max/Team/Enterprise (September 22, 2026): Anthropic also lifted five-hour usage caps on Pro/Max/Team/Enterprise seats and flagged Sonnet 5.5 and Haiku 5.5 for the coming weeks — the tier-consistent-pricing pattern from Claude Opus 5 holds while the usage-envelope loosens on the paid seats.
-
2026-09-29-AI-Digest — Claude Sonnet 5.5 ships Sept 28 with GDPval-AA v2.1 landing within two Elo of Claude Opus 5.5, and Terminal-Bench 4.0 jumping
10.3%→70.6%on the Sonnet tier at unchanged$2/$10per Mtok list price. Opus 5.5 surfaces as the ceiling comparator: the Sonnet-tier is now within measurement noise on GDPval-AA v2.1, which reframes the “when to default to Opus” calculus for tool-use-heavy agent stacks. Load-bearing softener: this is GDPval-AA v2.1 (agentic knowledge-work eval) — coding-specific benchmarks like SWE-Bench Pro have not been reported by Anthropic for Sonnet 5.5 yet, so the “collapse” is agentic-workflow-and-tool-use-scoped, not deep-coding-scoped. The Sonnet-5.5 free-tier claude.ai default deployment on the same day is what turns the capability delta into a practitioner-visible change. Reframe:Sonnet-tier is within 2 Elo of Opus on GDPval-AA v2.1 at Sonnet pricing, notAnthropic obsoleted Opus 5.5. The Sept-14 2026-09-15-AI-Digest flag that Sonnet 5.5 and Haiku 5.5 would ship “in the coming weeks” resolves as 2 weeks 6 days. Log against MOC - Major Companies and MOC - Agentic Coding. -
2026-10-02-AI-Digest — Claude Opus 5.5 surfaces as (1) comparator anchor in Cloudflare‘s Clef launch — Clef is sized and positioned for the guardrail / classification tier that sits in front of LLMs like Opus 5.5 or Gemini 4 Argon — and (2) as the first explicit mid-tier-step-up data point in today’s Key Takeaway on the convergent
$2/$10floor. Load-bearing framing to carry: Opus 5.5’s step-up curve plus Gemini 4 Argon‘s$4/$20standard rate are two independent data points suggesting the convergent-mid-tier$2/$10line will not hold through 2027 as a standard rate — intro periods will. Reframe:Opus 5.5 is one of two labs with published step-up curves; signal, not confirmation, notthe mid-tier floor has been abandoned. No fresh model-side action today; log as pricing-floor comparator anchor + control-plane-tier comparator. Log against MOC - Major Companies. -
2026-10-06-AI-Digest — Autonomous Claude Opus 5.5 agents proposed two room-temperature antiferromagnetic-semiconductor candidates — one new design, one
1999compound re-flagged — in a vals.ai writeup (HN front page,267pts /183cmts) that lays out the agent pipeline and verification steps. Load-bearing softener: a concrete materials-discovery claim from frontier agents in a domain with real device implications — tempered by the fact that one of the two is a reconfirmation, not a novel proposal. Opus 5.5 remains general-purpose reasoning substrate here (not a bespoke materials-science model); the agent-pipeline framing matters more than the model-choice framing. Log against MOC - Major Companies and MOC - Agentic Coding. -
2026-10-14-AI-Digest — Claude Opus 5.5 is the Opus-tier counterweight to Gemini 4 Argon‘s worldwide roll-out: Argon wins
12 of 18benchmarks on Google’s own table (DeepSWE 77.9 vs 74.2,Harvey Legal 19.6 vs 3.8) and Vals Index ranks Argon#1at68.90%vs Opus 5.5 at67.0%— but Opus 5.5 still leads onTerminal-bench 4.0(66.4 vs 57.4) andFrontierSWE v2(62.3 vs 55.0). Load-bearing softener: The Decoder’s “trailing” framing for Argon mixed categories by including Meta’s Muse Wonder (a consumer-agent frontier model, not a reasoning benchmark apex). Reframe worth carrying:Argon trades wins with Opus 5.5, Opus 5.5 is not displaced, notOpus 5.5 overtaken by Argon. Aider polyglot top-5 (fetched 2026-10-14) still carries no Opus 5.5 entry — measurement lag from the Sept-22 launch continues. Watch: a non-Google / non-Vals third-party benchmark pass this week before resetting the ceiling calculus. Log against MOC - Major Companies. -
2026-10-24-AI-Digest — Opus 5.5 surfaces as the Intelligence-Index and pricing-tier anchor against which Moonshot AI‘s new Kimi K3.1 Pro is measured — Artificial Analysis places K3.1 Pro at Intelligence Index
55, three points behind Opus 5.5’s58, with K3.1 Pro’s API at$2/$10per Mtok roughly half the implied Opus 5.5 rate (~$4/$20per Willison’s inference). Load-bearing framing: this is a comparator surfacing, not a fresh Opus 5.5 cut — no new release, benchmark pass, or product change today. Honest read the digest carries: K3.1 Pro undercuts on price while trailing on reasoning benchmarks (prior comparisons put K3 at56on HLE vs Opus’s64.7); Opus 5.5’s three-point Intelligence-Index lead holds, but the pricing-tier gap is what Moonshot is pressing. Log against MOC - Open Source Models (as comparator datum). -
2026-10-29-AI-Digest — Opus 5.5 surfaces as the agentic-coding ceiling that Google DeepMind‘s Oct 28 Gemini 4 Carbon still trails — per The Decoder, Carbon’s AAII v5 of
59is a nine-point jump over Argon’s50, and Carbon leads15 of 16of Google’s own benchmarks, but Google concedes Carbon still trails Opus 5.5 on agentic coding specifically. Load-bearing softener: specific numbers trace to a single secondary outlet (blog.googlenot independently reachable this run) — carry as The Decoder’s framing. No fresh Opus 5.5 release today; Aider polyglot top-5 unchanged with no Opus 5.5 entry. Reframe worth carrying: Carbon moves Google’s index-tier floor up while leaving Opus 5.5’s agentic-coding ceiling intact, not an Opus 5.5 displacement. Log against MOC - Major Companies (as comparator anchor). -
2026-11-02-AI-Digest — Opus 5.5 is the explicit migration target for Claude Opus 4.8 fast-mode workloads ahead of the
2026-11-30hard removal — the carried frame from Claude Codev2.1.316holds unchanged on Day 2 of the stability window. No fresh Opus 5.5 release today; log as cadence-carry — Opus 5.5 continues to pull Opus 4.8’s fast-mode demand on the migration side of the scheduled SKU compression. Log against MOC - Agentic Coding. -
2026-11-13-AI-Digest — Opus 5.5 holds the top of ExpertCrowd-SWE’s
100-task expert-authored multi-file benchmark on unfamiliar repos at67.3%, statistically tied with GPT-6 Astra at63.7%. Reframe worth carrying: the “plateau at~65%” framing is specific to hard, unfamiliar-codebase evals — on Aider-style familiar-language tasks frontier models still sit at85–88%. Carry as “saturation on easy,~65%ceiling on hard — not one plateau, two regimes”. Opus 5.5 remains unscored on the public Aider polyglot leaderboard (unchanged from yesterday’s fetch —gpt-5 (high/medium/low)still holds three of five slots at88.0/86.7/81.3%) so the measurement-lag framing continues on familiar-language tasks. No fresh Opus 5.5 release today; this is comparator-anchor carry on a regime-split reframe, not a model cut. Log against MOC - Agentic Coding and MOC - Developer Tools. -
2026-11-15-AI-Digest — Anthropic published direct practitioner guidance on prompting Claude Opus 5.5, explicitly flagging that effort setting — not phrases like “think carefully” — should be the primary control, that Opus 5.5 at Medium effort matches Opus 5 at High, and that starting at High is often wasteful. Other outdated habits called out: step-by-step instructions where a single clear goal would do, transcribing charts into text by hand rather than passing the image directly, asking for drafts before finished work, and over-specific negative constraints (“no generic design” and similar). The docs page is the primary source; The Decoder’s writeup is a legitimate derivation of first-party guidance. Carry as “effort setting is the main dial on Claude Opus 5.5; the ‘think hard’ / ‘start at High’ / ‘step-by-step’ habits are the ones Anthropic itself says cost the most against this model”. Log against MOC - Developer Tools and MOC - Agentic Coding.
Related
See also: Anthropic, Claude Fable 5.1, Claude Opus 5, Claude Code, GPT-6 Sol, GPT-6 Luna, MOC - Major Companies, MOC - Agentic Coding.