• Latest
  • Trending
  • All
Answer card: from July 20, 2026 Claude Fable 5 is included in every Max and Team Premium plan at 50 percent of normal limits, while Pro and Team Standard users get only metered access at ten dollars per million input and fifty dollars per million output tokens plus a one-time hundred dollar credit.

Claude Fable 5 goes permanent July 20: the fine print

18 July 2026
Answer card stating that Qwen-Image-2.1, released on 20 September 2026, ships open weights with a 7 billion parameter diffusion transformer, a Qwen3-VL 8B text encoder and an RGBA VAE totalling about 33 gigabytes in BF16, under the Qwen Research License that limits use to research or evaluation and requires a separate commercial licence, unlike the Apache 2.0 licence of Qwen-Image 1.0.

Qwen-Image-2.1 brings the weights back, but not the Apache licence

21 September 2026
Answer card stating that Ternary Bonsai 2 27B, released by PrismML on 17 September 2026 under Apache 2.0, packs Qwen3.8 27B into 5.95 gigabytes at 1.72 bits per weight, keeps 98.2 percent of the 14-benchmark average, about 75 percent on SWE-bench Verified and Terminal-Bench 2.1, and needs PrismML's llama.cpp fork to run.

Does Bonsai 2 27B really keep 98% of Qwen3.8 in 5.95 GB?

20 September 2026
Answer card stating that Jev 1.13 from TypeSafe AI is a decision model in early access since 15 September 2026 that returns typed probabilities instead of text, priced at 42 dollars per billion input tokens with output tokens free, answering in 70 to 500 milliseconds, with a 64K token request budget, text input only, and a documented list of things it does badly, including counting and dates.

Jev 1.13 bills $42 a billion tokens, and it can’t count

19 September 2026
Answer card stating that Qwen3.8-Omni-Flash launched on 17 September 2026 as an API only model on Alibaba Cloud Model Studio, taking text, images, audio and video in a 1M token context and returning text only, priced at 0.15 dollars per million input tokens for every modality and 0.47 dollars per million output tokens in the international regions, with no open weights published and the Qwen-Live Harness GitHub repository returning 404.

Qwen3.8-Omni-Flash bills audio at $0.15 and ships no weights

18 September 2026
Answer card stating that on 15 September 2026 AWS said it is unable to restore access to resources and data hosted exclusively in the Middle East Bahrain region me-south-1 and in the mec1-az2 zone of the UAE region, because the damage spanned multiple Availability Zones and exceeded what multi-AZ services are designed to withstand.

AWS can’t restore me-south-1, six months after the drone strikes

17 September 2026
Answer card stating that Google released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on 15 September 2026 at 3 dollars per million audio input tokens and 12 dollars out, that the thinking model requires asynchronous tools, and that Artificial Analysis scores it 82.6 on its Speech to Speech Quality Index.

Gemini 3.8 Live Extended Thinking rejects any tool that blocks

16 September 2026
Answer card summarising the Atria Dawn Preview release: 744B GLM-5.2 base, MIT licence, 1.5 TB BF16 and 756 GB FP8 checkpoints, 256K context, top on five of sixteen benchmark rows and trailing on SWE-bench Pro.

Atria Dawn Preview is 744B under MIT, and the BF16 weighs 1.5 TB

15 September 2026
Answer card stating that OpenAI released the Agents API in public beta on 10 September 2026 with no separate fee, billed through model tokens, tool calls and hosted sandbox time, with a choice of OpenAI hosted, self hosted or partner sandboxes, US only data residency and no Zero Data Retention support.

OpenAI’s Agents API has no fee, no ZDR and a one hour sandbox clock

14 September 2026
Answer card: Sakana Fugu Max at $2 and $6 per million tokens, Fugu Ultra v2 unchanged at $5 and $30, and Sakana saying Ultra v2 scores without Fable 5 or GPT-6 Astra in its pool.

Fugu Max costs $2 and $6 while Fugu Ultra v2 runs without Fable 5

13 September 2026
Answer card stating that DeepSeek released DeepSeek-V4.1-Flash on 10 September 2026 as a 552 billion parameter mixture of experts model with a new causal encoder decoder architecture that activates 8 billion parameters on input and 16 billion on output, with native vision, a one million token context and MIT licensed weights, that the API model name is now deepseek-flash at 0.15 dollars per million input tokens and 0.60 dollars per million output tokens off peak, and that DeepSeek announced V4 Pro would be routed to V4.1-Flash from 14 September and reversed that on 11 September.

DeepSeek V4.1-Flash arrived, and the V4 Pro retirement lasted a day

12 September 2026
Answer card stating that Cognition released SWE-2 on 10 September 2026, a coding model post-trained from Kimi K3, scoring 50.0 percent on FrontierCode 1.1 Main against 50.9 percent for Claude Fable 5.1 and 27.3 percent on Terminal-Bench 4 against 55.8 percent, available only inside Devin.

SWE-2 trails Fable 5.1 by one point, and by 28 on Terminal-Bench 4

11 September 2026
Answer card for Meta Muse, free to 100 million tokens a week then $20 a month, launched 8 September 2026 for United States adults only, running in a dedicated per user virtual machine.

Does Meta Muse do enough to earn your inbox and a card on file?

9 September 2026
  • About
  • Contact
  • Privacy
  • Legal
Monday, September 21, 2026
  • Login
Packet Nebula
  • Home
  • Articles
    • Security
    • Network
    • Dev
    • Sysadmin
    • SEO
    • Email & DNS
  • Tools
    • Network tools: free, fast, no signup
    • Security tools: free, fast, no signup
    • Developer tools: free, fast, no signup
    • Sysadmin tools: free, fast, no signup
    • SEO tools: free, fast, no signup
    • Email & DNS tools: free, fast, no signup
  • Download
  • About
No Result
View All Result
Packet Nebula
No Result
View All Result
Home Dev

Claude Fable 5 goes permanent July 20: the fine print

by stephane
18 July 2026
in Dev
0
Answer card: from July 20, 2026 Claude Fable 5 is included in every Max and Team Premium plan at 50 percent of normal limits, while Pro and Team Standard users get only metered access at ten dollars per million input and fifty dollars per million output tokens plus a one-time hundred dollar credit.
491
SHARES
1.4k
VIEWS
Share on FacebookShare on Twitter

You opened Claude Code this morning half expecting Fable 5 to be gone again, because the cutoff has slipped three times now. It isn't going anywhere. On July 18 Anthropic said Fable 5 becomes a permanent part of its subscriptions, and from July 20 it's included in every Max and Team Premium plan, capped at 50% of your usual limits. That much is real, and it's good news if you're on the right plan. Here's the part the headline skips: on the $20 Pro plan or Team Standard, permanent doesn't mean in your plan. You get metered access at the priciest rate Anthropic sells, softened by a one-time $100 credit that burns faster than you'd guess. We went and split the who-gets-it-free from the who-pays-by-the-token, because that line is the only bit that touches your bill.

The short answer

Anthropic is making Claude Fable 5 a permanent subscription option, ending the run of moving deadlines. From July 20 it’s included in every Max and Team Premium plan, but only at half your normal limits. Pro and Team Standard don’t get it included at all. They pay metered credits at $10/$50 a million, with a one-time $100 credit to soften the landing. Here’s who’s in and who’s billed.

July 20permanent, from
50%of limits, Max and Premium
$10 / $50per Mtok if metered
Answer card: from July 20, 2026 Claude Fable 5 is included in every Max and Team Premium plan at 50 percent of normal limits, while Pro and Team Standard users get only metered access at ten dollars per million input and fifty dollars per million output tokens plus a one-time hundred dollar credit.
The one-card version. Permanent, but included only on Max and Team Premium, and only at half your limits.

The announcement, minus the hype

Here’s the actual news. Anthropic said on July 18 that Claude Fable 5 stops being a temporary promo and becomes a standard part of its subscription lineup. Starting July 20, it’s included in all Max and Team Premium plans. The quote doing the rounds is short and worth reading literally: “Beginning July 20, Claude Fable 5 will be included in all Max and Team Premium plans, at 50% of limits.”

Read that last clause twice. Included, yes. At 50% of limits. So even on the plans that get it, you’re working with half your normal weekly allowance for Fable 5 specifically. That’s not a footnote. If you’re the kind of person who leaves a coding agent running for hours, you’ll feel the cap.

And the plans that don’t get it? That’s most subscribers. The $20 Pro plan doesn’t include Fable 5. Neither does Team Standard. You keep access on those, but only through metered usage credits, billed per token like the raw API.

Included, or metered? Find your line

This is the whole practical question, so let’s be blunt about it.

If you’re on Max, the $100 or $200 a month tier, Fable 5 is included from July 20 at 50% of your limits. If you’re on Team Premium, same deal. Those two are the winners here.

If you’re on Pro at $20 a month, or Team Standard, Fable 5 is not part of your allowance. You reach it through usage credits, and those bill at $10 per million input tokens and $50 per million output. That output number is the one that stings. It’s the most expensive rate Anthropic publishes for a general model, and Fable 5 is a reasoning-heavy model that likes to produce a lot of output tokens.

To take the edge off the switch, Pro and Team Standard users get a one-time $100 credit. Handy, but understand what it is. A cushion, not a plan.

Checklist of who gets Claude Fable 5 included versus who pays metered from July 20 2026: Max and Team Premium include it at 50 percent of limits, while Pro at twenty dollars a month and Team Standard get metered access only at ten and fifty dollars per million tokens, both softened by a one-time hundred dollar credit.
Two plans include it. Everyone else is on the meter. Find your row before you plan a build.

About that $100 credit

Let’s do the arithmetic nobody puts in the announcement. The one-time credit is $100. Metered output runs $50 a million tokens. So the credit buys you somewhere around 2 million output tokens, plus whatever your input adds on top at $10 a million.

Two million tokens sounds like a lot until you’ve watched an agentic session work. Fable 5 in a Claude Code loop, chewing through a refactor across a dozen files, rereading its own output, retrying: that burns output fast. I’d budget the $100 as maybe a few solid days of heavy use, not a month of it. Your mileage genuinely varies here, so watch the meter for your first week before you assume.

The honest read: if Fable 5 is a daily driver for you and you’re on Pro, that $100 is a countdown, and Max starts to look like the cheaper answer. If you dip into Fable 5 now and then for the hard problems and run Sonnet 5 or something lighter for the rest, the metered credits on Pro will almost certainly beat paying five times as much for a Max seat.

Why the drama, and why now

The reason it took this long, and this many deadline slips, is capacity. Anthropic has been straight about it: demand for Fable 5 is “very high, and difficult to predict.” So rather than promise the model into every plan and then throttle everyone when the servers filled, it ran a conservative rollout and kept the cutoff on a short leash.

That leash moved a lot. Fable 5 came back into subscriptions on July 1 at up to 50% of weekly limits, set to drop to credits-only after July 7. Then July 7 arrived and the deadline jumped to July 12. On July 13 it jumped again to July 19. The July 18 post is where the pattern finally breaks: instead of another week’s extension, Anthropic committed to keeping Fable 5 as a permanent fixture on Max and Team Premium. Same 50% cap it’s had all along, now with no expiry date attached.

Timeline of Claude Fable 5 subscription access in July 2026: it returned on July 1 included through July 7, the cutoff was pushed to July 12 on July 7, pushed again to July 19 on July 13, and on July 18 Anthropic made it permanent in Max and Team Premium from July 20 at 50 percent of limits.
Three extensions, then a decision. July 20 is where the model stops having an expiry date.

What to actually do

Check which plan you’re on. That’s step one, and it decides everything else.

On Max or Team Premium, you’re set. Fable 5 is yours from July 20 at half limits, and you don’t have to think about tokens. Just know the cap exists so a long agent run doesn’t surprise you.

On Pro or Team Standard, decide whether Fable 5 is a habit or a treat. If it’s a habit, price out a Max upgrade against what $50-a-million output will cost you at your real usage, and lean on the $100 credit to measure that before you commit. If it’s a treat, stay on Pro, keep Fable 5 for the problems that actually need it, and let a cheaper model carry the routine work. Either way, turn on a spend cap if the option’s there, because metered billing at $50 out has no brakes of its own.

And whatever the marketing says, treat “permanent” as “permanent on two plans, at half limits.” That’s the accurate version, and it’s still a decent outcome. The model that a lot of us reached for the moment it came back isn’t getting yanked again. It just isn’t free for most of us.

Sources: Anthropic (Redeploying Fable 5), Simon Willison, BleepingComputer and DigitalApplied (pricing), July 2026. The permanent inclusion in Max and Team Premium from July 20 at 50% of limits, the metered $10 per million input and $50 per million output pricing, and the one-time $100 credit for Pro and Team Standard are per Anthropic’s own communications as reported. The $100-buys-about-2M-output-tokens figure is our own arithmetic from the published output rate, not an Anthropic statement, and real usage varies.

Frequently asked questions

Is Claude Fable 5 permanent now?

Yes, as a subscription option. On July 18, 2026 Anthropic said Fable 5 becomes a standard, permanent part of its plans rather than a rolling promo. Starting July 20 it is included in every Max and Team Premium plan at 50% of your normal usage limits. That ends the run of temporary extensions that kept moving the cutoff.

Which Claude plans include Fable 5?

Only Max (the $100 to $200 a month tiers) and Team Premium include Fable 5 from July 20, and only up to 50% of your weekly limits. The $20 Pro plan and Team Standard do not include it. On those you keep access, but through metered usage credits billed per token.

How much does Fable 5 cost if my plan does not include it?

It bills at $10 per million input tokens and $50 per million output tokens, the most expensive rate Anthropic lists for a general model. Pro and Team Standard users get a one-time $100 credit to cushion the move to metered billing. At $50 a million out, that credit covers roughly 2 million output tokens, which a heavy agentic-coding day can eat through.

Why did Anthropic keep extending the Fable 5 deadline?

Capacity. Anthropic said demand for Fable 5 is very high and hard to predict, so it kept the model on a conservative rollout and pushed the subscription cutoff from July 7 to July 12, then to July 19, while it added serving capacity. The July 18 announcement is the point where it decided it could commit to keeping the model in plans for good.

Should I upgrade to Max just to get Fable 5?

Only if you actually reach for Fable 5 a lot. Max runs $100 to $200 a month and gives you Fable 5 included at half your limits, which beats paying $50 a million out on Pro if you are a daily heavy user. If you touch Fable 5 occasionally, the metered credits on Pro will almost certainly cost you less than the Max jump.

Tags: aianthropicclaudellmnewspricing
Share196Tweet123
stephane

stephane

  • Trending
  • Comments
  • Latest
Answer card: Proton Lumo 2.0 is private by policy, not by locality. Saved history is locked so even Proton cannot read it, but the prompt is decrypted on a Proton EU server to answer it, then forgotten.

Proton Lumo 2.0 review: how private is it, really?

3 September 2026
The Agentic Coding section of the official Hy4 preview benchmark appendix published by Tencent, a table comparing Hy3 and Hy4 preview against DeepSeek V4 Pro 0813, Qwen 3.8 Max, GLM 5.3, Kimi K3, GPT 5.6 Sol and Claude Opus 5 across SWE-bench Multilingual, SWE-bench Pro, DeepSWE, three SWE Atlas tasks, SWE-Marathon, Terminal-Bench 2.1, NL2Repo-Bench, CyberGym, ProgramBench, PostTrainBench and Harbor-Index.

Tencent’s 770B Hy4 tops one benchmark row in 46

3 September 2026
Answer card: Qwen 3.7 Max is API-only and cannot run locally yet; the open Qwen models (Qwen 3.6 27B, qwen3:8b to 32b) run offline via Ollama.

Qwen 3.7 local: what you can actually run offline

22 June 2026
Answer card: JWTs are not encrypted, anyone can read them; the signature proves who issued the token, not who may read it.

Are JWTs encrypted? No, and the difference will bite you

0
Answer card: a random 8 character password falls in under 2 hours offline, while 16 random characters hold for 1.4 trillion years at the same speed.

How long does it take to crack a password in 2026?

0
Answer card: three DNS records decide if your mail lands or bounces; SPF lists allowed senders, DKIM signs messages, DMARC sets the failure policy.

SPF, DKIM and DMARC explained: the records your email needs

0
Answer card stating that Qwen-Image-2.1, released on 20 September 2026, ships open weights with a 7 billion parameter diffusion transformer, a Qwen3-VL 8B text encoder and an RGBA VAE totalling about 33 gigabytes in BF16, under the Qwen Research License that limits use to research or evaluation and requires a separate commercial licence, unlike the Apache 2.0 licence of Qwen-Image 1.0.

Qwen-Image-2.1 brings the weights back, but not the Apache licence

21 September 2026
Answer card stating that Ternary Bonsai 2 27B, released by PrismML on 17 September 2026 under Apache 2.0, packs Qwen3.8 27B into 5.95 gigabytes at 1.72 bits per weight, keeps 98.2 percent of the 14-benchmark average, about 75 percent on SWE-bench Verified and Terminal-Bench 2.1, and needs PrismML's llama.cpp fork to run.

Does Bonsai 2 27B really keep 98% of Qwen3.8 in 5.95 GB?

20 September 2026
Answer card stating that Jev 1.13 from TypeSafe AI is a decision model in early access since 15 September 2026 that returns typed probabilities instead of text, priced at 42 dollars per billion input tokens with output tokens free, answering in 70 to 500 milliseconds, with a 64K token request budget, text input only, and a documented list of things it does badly, including counting and dates.

Jev 1.13 bills $42 a billion tokens, and it can’t count

19 September 2026
  • About
  • Contact
  • Privacy
  • Legal

Copyright © 2026 Stephane Cardon.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Home
  • Articles
    • Security
    • Network
    • Dev
    • Sysadmin
    • SEO
    • Email & DNS
  • Tools
    • Network tools: free, fast, no signup
    • Security tools: free, fast, no signup
    • Developer tools: free, fast, no signup
    • Sysadmin tools: free, fast, no signup
    • SEO tools: free, fast, no signup
    • Email & DNS tools: free, fast, no signup
  • Download
  • About

Copyright © 2026 Stephane Cardon.