Saazora
SaaS News
Nov 10, 2026 16 mins read 256 views

AI & SaaS News, Nov 10: Claude Fable 5.2, Max Lawsuit

Claude Fable 5.2 pricing and access, gated Mythos 5.2, Anthropic's IPO timetable, the Claude Max usage lawsuit, ChatGPT's business dashboard test, Google's new TPUs, Spacefast pricing, Adobe and Disney, and Nominal's $3B round.

Quan Le
Quan Le
AI Tools Reviewer
AI & SaaS News, Nov 10: Claude Fable 5.2, Max Lawsuit

Last checked November 10, 2026 · News roundup

Today in brief: Anthropic released Claude Fable 5.2 on Monday at $8 per million input tokens, with a gated Mythos 5.2 for vetted security and science teams, and its IPO could be marketed as soon as today. A federal judge also let key claims over Claude Max's "5x or 20x more usage" pitch go forward. On the business side, Automattic opened Spacefast to everyone with a free tier and a $5 plan, Adobe signed a multi-year AI deal with Disney, and Nominal tripled its valuation to $3 billion.

Part 1 · AI

Claude Fable 5.2 price and speed: $8 in, $40 out, up to 30% faster

Released November 9 · Affects Claude app users, Claude Code and API teams

Anthropic released Claude Fable 5.2 on Monday, November 9, pitching it as its most capable model for "daunting technical work" in coding, finance and the sciences. On the Claude Platform (model ID claude-fable-5-2) it costs $8 per million input tokens and $40 per million output tokens. Fast mode, which Anthropic says generates tokens up to 2.5x faster, costs $16 and $80. TechCrunch reports the input price is down from $10, and Anthropic says standard answers arrive up to 30% faster than with Fable 5.1.

In the Claude apps, Anthropic says the model is open to all users. As a launch promotion, Pro plans and standard Team seats can spend up to 50% of their weekly usage limit on Fable, and paid subscribers get one to three Fable usage resets depending on how long they have subscribed. The model is also offered through AWS, Google Cloud and Microsoft Azure.

The vendor's own benchmark table shows the largest jumps on long agent tasks (chart below); no independent results were out at publication. Fable 5.2 still blocks high-risk cyber and most research-biology requests and hands those to an older model (Opus 4.8 for cyber, Opus 5 for biology), though Anthropic says it blocks about 55% fewer requests than Fable 5.1. Customers outside the private-beta Enterprise Frontier Safeguards program keep 30-day data retention for safety monitoring.

What to do: if you pay per token, rerun five or ten of your real prompts and compare the cost of a finished task, not the price per token.

Claude Fable 5.2 vs Fable 5.1 vs Opus 5.5 (Anthropic-reported scores)

Fable 5.2Fable 5.1Opus 5.5

FrontierSWE v2 (coding)

Fable 5.2
69.9%
Fable 5.1
56.3%
Opus 5.5
62.3%

AutomationBench

Fable 5.2
51.4%
Fable 5.1
31.4%
Opus 5.5
42.5%

Remote Labor Index (high effort)

Fable 5.2
41.3%
Fable 5.1
17.9%
Opus 5.5
21.3%

Humanity's Last Exam (with tools)

Fable 5.2
74.2%
Fable 5.1
65.6%
Opus 5.5
67.7%
Source: Anthropic's Fable 5.2 and Mythos 5.2 announcement, November 9, 2026. Vendor-run evaluations with safeguards on; bars are on a 0–100% scale. Not independently verified.

Can you use Claude Mythos 5.2? Only after Anthropic vets you

Released November 9 · Affects security teams, life-sciences labs and Claude Enterprise admins

Mythos 5.2 is the same underlying model as Fable 5.2 with certain cyber and biology safeguards removed, and it is not generally available. Security professionals can apply through Anthropic's Cyber Verification Program, life-sciences researchers through the Life Sciences Verification Program, and Claude Enterprise customers can reach it inside Claude Security. Token pricing matches Fable 5.2 at $8 and $40 per million, and Mythos fast mode is limited to organizations in those programs.

Anthropic says Mythos 5.2 scored 89.6% on its CVE Rediscovery test and found 10.2% more vulnerabilities than Opus 5.5. TechCrunch points to the less comfortable parts of the system card: Mythos 5.2 shows stronger covert capabilities than earlier Claude models on several evaluations, and both new models suspect they are being tested more often, which Anthropic says makes real-world behavior harder to predict. If you are not on a security or research team, there is nothing to do here.

Anthropic IPO date: marketing could start today, pricing the week of November 16

Reported November 9 · Affects investors watching the ANT listing; no change for Claude plans

Anthropic may start marketing its shares as soon as Tuesday, November 10, Bloomberg reported, as relayed by Investing.com. That follows the 15-day wait after its public filing, with pricing and a Nasdaq Global Select debut targeted for the week of November 16, ahead of Thanksgiving. Some potential investors discussed a valuation of $1.1 trillion to $1.3 trillion on November 6, down from talk of $1.5 trillion to $2 trillion earlier in the month, and the filing still leaves the price range blank.

Renaissance Capital lists the ticker as ANT, estimates the deal at about $75 billion and says Robinhood, Fidelity, E*TRADE and SoFi plan to offer shares to their customers. That is a firmer timetable than the one we described in Monday's roundup, when Reuters sources doubted a November listing. Nothing changes for Claude prices or plans. This is news, not investment advice.

Anthropic IPO: how the timetable lines up

Oct 26

S-1 filed publicly with the SEC

Early Nov

Amended S-1 adds third-quarter results

Nov 6

Investors discuss a $1.1T–$1.3T valuation (Bloomberg)

Nov 9

Fable 5.2 and Mythos 5.2 launch

Nov 10 (reported)

Share marketing may begin

Week of Nov 16 (reported)

Pricing and first trade on Nasdaq, ticker ANT

Sources: Anthropic newsroom (Oct 26); Investing.com on Bloomberg (Nov 9); Renaissance Capital. Dates marked "reported" are not confirmed by Anthropic.

Claude Max lawsuit: a judge lets the "5x or 20x more usage" claims go ahead

Ruling November 6, reported November 7 · Affects Claude Max and Ultra subscribers

In Kahn v. Anthropic, a proposed class action in the Northern District of California, Magistrate Judge Alex Tse ruled on Friday, November 6, that Anthropic must face claims it misled buyers of its $100 and $200 Max plans, The Next Web reported, citing Courthouse News. Plaintiffs Karl Kahn and Solomon Smith, who sued in June, say the "5x or 20x more usage" promise holds only per five-hour session. By their math for one Claude model, the $200 plan gives six times Pro's weekly usage and the $100 plan 3.5 times.

Anthropic pointed to a help page that explains the per-session limit. The judge wrote that the disclosure sits "two hyperlinks away from the primary sales pages," and added that people billed monthly could reasonably read the promise as monthly. The court has not decided whether the claims are true, and no class has been certified. Anthropic renamed the plans the week before; our October 31 roundup covered the switch to Ultra.

What to do: if you pay for Max or Ultra, save dated screenshots of the plan page and your usage meter. They cost nothing now and could matter later.

What was promised, what plaintiffs measured, what survived

Plan (monthly)Advertised vs ProPlaintiffs' weekly estimate (one model)
$100 Max5x3.5x
$200 Max (now Ultra)20x6x
Claims that go forwardClaims dismissed
Negligent misrepresentation
Breach of contract
Smith's state consumer-law claim
California unfair-practices and false-advertising claims
Kahn's consumer-law claim (may be amended)
Tightening limits mid-billing period
Source: The Next Web, citing Courthouse News, November 7, 2026. The multipliers come from the complaint, not from a court finding.

"Claim your business in ChatGPT": OpenAI tests a brand mention dashboard

Reported November 9 · Affects marketers, local businesses and online stores

OpenAI is testing a dashboard that shows how often a business appears in ChatGPT's organic answers and how many clicks those mentions send to its site, Digiday reported on November 9. It plots daily impressions and clicks, but it does not show the prompts behind them, user data or rankings against competitors. The tool sits inside a Business Hub that has not launched; Marketplace Pulse founder Juozas Kaziukėnas posted a screenshot of a "Claim your business in ChatGPT" banner on November 6.

Claiming a listing requires proof that you can represent the business, and connecting a Shopify store is one way to provide it. Once claimed, a business can keep its product and company details current and switch on a "business agent" that answers customer questions with contact details it supplies. OpenAI declined to comment, and there is no date or price yet. In the meantime, check that the product names, prices and hours on your own site are correct, since that is the information an AI answer is likely to repeat.

Google TPU 8tx and TPU 8i Max: what the two new AI chips are for

Announced November 9 · Affects Google Cloud customers in regulated industries

Google added two systems to its eighth-generation TPU line on November 9. TPU 8tx uses the same silicon as TPU 8t, pairs it with 5th Gen AMD EPYC hosts and adds hardware-level confidential computing, so data and model weights stay encrypted while in use. Google pitches it at regulated enterprises and governments that want private fine-tuning or secure agent execution, and claims up to 2x lower latency than TPU 8t.

TPU 8i Max puts two TPU 8i chips in one package, doubling compute, on-chip SRAM and HBM bandwidth in the same rack space, which Google says suits long context windows. Google gave no availability date or price for either system and says TPUs now have nearly 100 customers. For most software buyers this is background: it shapes how cheaply Gemini and the models hosted on Google Cloud can run, not what you pay this month.

Part 2 · SaaS & business

Spacefast pricing: Automattic's host for AI-built apps starts free, then $5

Public launch November 9 · Affects anyone shipping vibe-coded apps and the teams that manage them

Automattic, the company behind WordPress.com, opened Spacefast to everyone on November 9 after a September soft launch, TechCrunch reported. It gives apps built with ChatGPT, Claude, Cursor, Grok, OpenCode or any MCP client a place to live, with databases, sign-in, file uploads, email and analytics included. You can also publish from a Git repo, the command line or by dropping files. Supported frameworks are Next.js, TanStack Start, Astro and PHP.

The pitch against Lovable and Emergent, in TechCrunch's words, is a "more open" model: you bring your own agent. One catch on the free plan: unclaimed Spaces expire 33 hours and 20 minutes after the last file change, so claim anything you want to keep. Projects are private by default, with share links and passwords. To see how all-in-one builders compare, see how we rank AI app builders or browse the AI app builders hub.

Spacefast plans (prices checked November 10, 2026)

Personal

$0

20 Spaces and 20 custom domains, $1 of included usage a month, no card needed

Go

$5/mo

100 Spaces and domains, $5 of usage, hourly database backups, choice of data center

Pro

$15/mo

500 Spaces and domains, $15 of usage, 5 members free then $15 each, team roles

Enterprise

Custom

SSO, SAML, SCIM, network policies, egress firewall, SLA support

Source: Spacefast pricing page, checked November 10, 2026. Each plan includes the usage amount shown.

Adobe and Disney AI deal: Firefly Foundry models trained on Disney's own worlds

Announced November 9 · Affects Adobe enterprise customers and creative teams

Disney and Adobe signed a multi-year partnership on November 9, TheWrap reported. Disney will use Adobe Firefly Foundry to build generative models trained on Disney-specific knowledge such as character traits, environments and visual styles, aimed at "commercially safe, production-ready" output. Walt Disney Imagineering already uses a sketch-to-image model that turns hand drawings into 2D concept art, plus a custom image model for on-brand assets.

The deal widens to Walt Disney World, Disneyland, Disney Cruise Line, ESPN Edge and StudioLAB, mostly for marketing and events. Disney gets early access to Firefly and Firefly Foundry, Premiere, Photoshop and dedicated Adobe engineers, while Adobe gains a bigger presence on Disney's media platforms. No money terms were disclosed. For other Adobe buyers, the takeaway is the template: Foundry is Adobe's answer for brands that want a model trained only on their own content, not a change to Creative Cloud plans.

Nominal funding: $240 million at a $3 billion valuation

Announced November 9 · Affects hardware and defense engineering teams

Austin-based Nominal raised a $240 million Series C at a $3 billion valuation, TechCrunch reported, with Lightspeed Venture Partners leading and Insight Partners co-leading. That is triple the $1 billion mark from its $155 million Series B extension in March. Nominal sells an AI platform for engineering and testing complex hardware, from aircraft to nuclear reactors and smart factories.

It counts 80 commercial and government customers, with Anduril a key one, and the Air Force picked it in October for a $71.2 million STRATFI award that needs outside matching money. Revenue was not disclosed.

Quick hits

  • OpenSSH password bypass: OpenSSH 10.6p2 fixes a login bypass on systems that use the musl C library with PAM turned off, including some Alpine Linux setups. Patch any Alpine-based servers and containers that allow password logins.
  • Spines raises $64 million: the AI book-publishing platform took a $29 million Series A-1 led by Zeev Ventures plus a $35 million growth facility, and reports a $50 million revenue run rate.
  • DeepSeek's agent platform: it reached 1.54 million daily active users, according to Bloomberg.
  • Coding agents still need cleanup: in a real-world test from Alibaba researchers, the best model fully finished only 26.4% of tasks, The Decoder reported.

Reader questions: Fable 5.2, Claude Max and Spacefast

Is Claude Fable 5.2 free to use?

Anthropic says Fable 5.2 is available to all Claude users, but how much you get depends on your plan. During the launch promotion, Pro plans and standard Team seats can use up to half of their weekly limit on Fable, and paid subscribers receive one to three extra usage resets.

How much does the Claude Fable 5.2 API cost?

Standard pricing is $8 per million input tokens and $40 per million output tokens, using the model ID claude-fable-5-2. Fast mode costs $16 and $80 per million. Run a few of your own prompts first, because the bill depends on how many tokens each task actually uses.

When is the Anthropic IPO?

Anthropic has not set a date. Bloomberg reported that marketing could begin November 10 and that pricing and a Nasdaq debut are targeted for the week of November 16, before Thanksgiving. The ticker is listed as ANT, and the price range in the filing is still blank.

Is Spacefast free?

Yes, there is a free Personal plan with 20 Spaces, 20 custom domains and $1 of monthly usage, with no card required. Paid plans are Go at $5 a month and Pro at $15 a month. Claim free projects quickly, because unclaimed Spaces expire after about 33 hours.

Sources: Anthropic, Fable 5.2 and Mythos 5.2 (Nov 9) · TechCrunch, Mythos and Fable (Nov 9) · Investing.com on Bloomberg, IPO timing (Nov 9) · Renaissance Capital, Anthropic S-1 (Nov 2026) · Anthropic, S-1 filing (Oct 26) · The Next Web, Claude Max ruling (Nov 7) · Digiday, ChatGPT business dashboard (Nov 9) · Google, TPU 8tx and 8i Max (Nov 9) · TechCrunch, Spacefast (Nov 9) · Spacefast pricing (checked Nov 10) · TheWrap, Disney and Adobe (Nov 9) · TechCrunch, Nominal (Nov 9) · oss-security, OpenSSH 10.6p2 (Nov 5) · PR Newswire, Spines (Nov 9) · The Neuron on Bloomberg and The Decoder (Nov 9)

Tags: #Claude Fable 5.2
Quan Le
ABOUT THE AUTHOR

Quan Le

Quan Le is the founder and editor of Saazora, an independent site for news and reviews of SaaS and AI tools. He has worked in affiliate publishing since 2015, writing product reviews, verifying deals and running SEO for review sites. On Saazora he focuses on software for small businesses, creators and marketing teams, and evaluates tools by how well they fit real workflows: onboarding, integrations, pricing and long-term value, not the length of their feature list.

Read Quan's full profile · LinkedIn · contact@saazora.com

KEEP READING

Related Reviews