Dan Shipper, CEO of Every, spent a week testing Anthropic's Fable 5.1 AI model. This hands-on "vibe check" review covers the model's performance on coding, knowledge work, and writing tasks, and closes with overall impressions. # Video Identity Observations: "We Tested Anthropic's Fable 5.1 for a Week" ## Explicit Self-Description The video contains a direct creator self-introduction segment: > "My name is Dan Shipper, and I'm the co-founder and CEO of Every. Every is the only subscription you need to stay at the edge of AI." - **Claimed access model:** "We get all of the models before they come out, and we test them with a team of writers, designers, engineers, and more to tell you what they're good for." - **Publishing cadence claim:** "On the day a model drops, we have long-form reviews to tell you what to use them for and what not to use them for from our real experience doing real work." - **Companion properties:** References to "every.to" (long-form written review with benchmarks) and a downloadable "prompt library" indicate a media/newsletter operation beyond the channel itself. ## Intended Audience The video addresses two overlapping audiences, evidenced by the framing of demonstrations: - **Technical/developer viewers:** Extended coverage of the "Hands" computer-use agent, "UltraCode" runs of "40 sub agents for like a day," token/latency benchmarks (766 tokens vs. ~2,000 for Opus 5; 22s vs. 37s latency), and pricing/token-efficiency comparisons. - **Knowledge workers:** Repeated emphasis that delegation is newly accessible: "That hasn't really been possible for knowledge workers… This is a big step in the direction of, actually, I can have it make a deck for me end to end." - The creator explicitly positions the model as reaching "more average knowledge workers" and viewers who might have thought "this is not for me." ## Claimed Scope - **Full-workflow coverage claims:** "We've done it on coding… knowledge work like decoration and data analysis… and writing." - **Cross-model comparative scope:** Direct head-to-head evaluations against "GPT 5.6" and "Opus 5" throughout (survey dashboards, PowerPoint decks, one-shot blog posts). - **Internal benchmarking:** References "our internal agent benchmark" measuring "tasks that our internal agent does like work inside of a company," implying an institutional, team-based testing operation rather than individual review. ## Observable Presentation Patterns - **Casual, first-person evaluative voice:** "This is a sick model… Fable all day, baby" — informal, opinionated commentary rather than neutral reporting. - **Transparency about bias and limits:** Acknowledges costs ("this is like a three to five million token run… it can get expensive") and personal preference ("I'm still using 5.6… for most of my day-to-day writing"). - **Named team members cited as evidence:** Kieran Klaassen, Mike Taylor, Katie Parrott — reinforcing a collective, staff-based evaluation brand. - **Show-your-work formatting:** Screenshots of usage graphs, prompts, eval outputs, and reading-ease comparisons. ## Creator Claims vs. Demonstrated Patterns - *Claim:* Every tests models "with a team" before release — *demonstrated* only by this video's single-author narration plus secondhand quotes from colleagues. - *Claim:* "Real experience doing real work" — *partially demonstrated* via usage graphs, an internal survey dataset, and real internal documents (meeting flagging, blog post generation), though all are creator-presented without independent verification. - *Caveat:* All benchmark numbers are self-reported; the transcript provides no methodology for the "internal agent benchmark." # Delivery Observation Report — *We Tested Anthropic's Fable 5.1 for a Week* ## Overall Impression An energetic, first-person product review delivered in a fast, conversational register. The speaker blends enthusiast slang with concrete numbers and live demos, moving through a fixed structure: intro → credibility → coding → knowledge work → writing → usage data → recommendation. ## Tone - **Enthusiastic and casual, with profanity used for emphasis:** > "This is a sick model." > "And that is like fucking wild. Look at that." > "It's like better for especially for big projects… but fucking sick." - **Self-aware showmanship at the top:** > "It's model release day! We've got an all-new Fable, Fable 5.1, and it is Fable for everyone." - **Balanced moments do occur**, though they are comparatively brief and hedged with casual qualifiers: "That said, there are some things to be aware of… So it can get it can get expensive." ## Recurring Phrases and Vocabulary - **"Slop"** — repeated as a quality verdict on model output: > "It doesn't feel like slop." / "It looks a little bit like slop." / "it doesn't have that slop feeling." - **"End to end"** — the most repeated compound, appearing across demos: > "built end to end by Fable 5.1 from a couple prompts" / "It created this deck end to end" / "it just built this app totally end to end." - **"It just works" / "it just worked":** > "And this actually just works." / "And then this came out and it just it just works." - **"Vibe check"** and **"vibe coding"** — framing vocabulary for informal testing. - **Comparative scaffolding:** near-constant explicit contrasts — "It's priced like a Fable, but you'll be able to use it for all your Opus tasks," "twice as fast as Opus," "It's like a little bit worse. It's a little bit more bare." - **Casual intensifiers and hedges:** "like," "I mean," "you know," "sort of," "a little bit." ## Teaching Habits - **Concrete numbers as proof points.** Benchmarks are quoted exactly rather than described qualitatively: > "on average, this model consumed about 766 tokens for each run… Opus 5 was almost 2,000 tokens" / "average latency for this model was about 22 seconds… Opus was about 37 seconds." - **Named comparative models throughout** — GPT 5.6 and Opus 5 outputs are shown side by side and critiqued, not just referenced. - **Defined jargon inline for a general audience:** > "NPS is net promoter score" - **Demo-first explanation:** he shows the artifact (the "Hands" window, the deck, the blog post) and narrates what to notice — "Look at that. It's look at that." / "You can see the arrows here. The arrows are not done right." - **Direct address and takeaway summaries:** "So the thing I want you to take away is developers for at least the last six months… have been getting used to delegating big chunks of work." - **Named colleagues as sourcing:** Kieran Klassen, Katie Parrott, Mike Taylor — attributing team-level conclusions to individuals. ## Communication Patterns - **Rhetorical question pivots to transition sections:** "So before we get into this video, who am I, and how do I possibly have a full review of a model that just dropped?" / "Why is that? That's interesting, right?" / "Okay. What did I actually use the tokens on?" - **Credibility insert early:** a personal intro block ("My name is Dan Shipper, and I'm the co-founder and CEO of Every…"). - **Formulaic closing recommendations addressed to audience segments:** "if you love the original Fable, you're going to love this model even more" and "if you like Claude models… you're going to love this." - **Explaining significance through historical analogy:** > "There was this moment in like December 2025, where Opus 4.5 and GPT 5.3-ish… cross this barrier… I think there's a similar thing happening with Fable 5.1." - **Imaginative framing devices:** the "McDonald's eval" and "I just let it cook" as shorthand for model behaviors. - **Spoken disfluencies retained** (repetitions like "it can get it can get expensive," "that that hasn't really been possible"), suggesting a largely unscripted delivery. ## Stable Patterns vs. Isolated Moments - **Well-established (repeated across sections):** "end to end," "slop," comparative benchmark quoting, rhetorical-question transitions, and demo-then-narrate structure all recur across the coding, knowledge work, and writing segments. - **Single-occurrence (report with caution):** the profanity appears only in the "Hands" demo; the "McDonald's eval" analogy, the "one-two punch" metaphor, and the historical December 2025 analogy each appear once. These are notable but should not be generalized as signature habits on this evidence. # Observation Record: "We Tested Anthropic's Fable 5.1 for a Week" ## Speaker Identification The transcript contains one identified speaker: **Dan Shipper**, who identifies himself as "the co-founder and CEO of Every." He cites colleagues' opinions at points (Kieran Klaassen, Katie Parrott, Mike Taylor) but those individuals do not speak in the transcript. - **Note on the transcript**: The speaker says "My name is Dan Shipper" and that he is reviewing "Anthropic's Fable 5.1." No curated metadata beyond the video title is supplied, so all positions below are attributed to Dan Shipper / "the speaker" unless explicitly attributed to a named third party in the transcript. --- ## Overall Position Shipper presents a strongly positive review of **Fable 5.1**, framed as a model that combines the delegation power of the original Fable with the cost/speed profile of smaller models. > "It's priced like a Fable, but you'll be able to use it for all your Opus tasks, and there's really no reason for you to use Opus anymore." --- ## Claimed Benchmarks and Metrics Attributed to Shipper's own testing at Every (unverified by this record): - **Token efficiency**: ~766 tokens per run on Every's internal agent benchmark vs. ~2,000 for Opus 5 ("uses about half the tokens"). - **Latency**: ~22 seconds average per request vs. ~37 seconds for Opus 5 ("about twice as fast"). - **Writing metrics**: He claims Fable 5.1 has "the highest reading ease and the lowest grade level of Opus 5 and GBT 5.6 Sol" (sic — "GBT 5.6 Sol" appears to be a transcription of a GPT model name; preserved as stated). - **Cost caveat (qualification)**: Large UltraCode runs "can get expensive" — he cites a 3–5 million token, roughly day-long run for the Hands app. --- ## Coding Claims - **Hands app**: Shipper claims he previously tried to build "Hands" (a computer-use agent controllable remotely, e.g., via Slack) in "5.6" and "regular Fable" and "it didn't really work," but Fable 5.1 "actually just works." He says he built it from "a couple prompts," running on UltraCode with "like 40 sub agents for like a day." - **Explicit qualification**: "I have no idea how this works." - He contrasts usage patterns: on UltraCode it's "not really that conversational" — it "goes off and cook[s] and come[s] back later," better for big projects than interactive coding. - **Colleague attribution**: He says Kieran Klaassen ("GM of Quora and makes compound engineering," a longtime "Claude stan") "didn't like Opus 5, didn't like Sonnet 5," and now calls Fable 5.1 "my favorite again." (Reported speech, not directly verified.) ## Historical Analogy (speculative claim, flagged by speaker) > "There was this moment in like December 2025, where Opus 4.5 and GPT 5.3-ish, like, cross this barrier... I think there's a similar thing happening with Fable 5.1." He suggests Fable 5.1 could "open up real delegation of coding and knowledge work tasks to... more average knowledge workers." This is framed as a prediction ("has the chance to"), not a settled fact. --- ## Knowledge Work Claims ### Data analysis - On an NPS dashboard task from a real customer survey, he says Fable 5.1 did "an incredibly good job at pulling out real insights from both the quantitative and qualitative data," e.g., inferring that "interesting" is "a warning word" while "useful, great, and smart predict promoter." - **Negative comparison preserved**: He says **GPT 5.6** did something better on this task — it "pulled out a story at the top" — and states, "I actually really like GPT 5.6 on this task." ### PowerPoint generation - From the prompt "take compounding engineering and turn it into a deck," he says Fable 5.1 produced a strong end-to-end deck, noting subtle successes like correctly drawn arrows/bubbles that "most models can't get." - **Small flaw preserved**: "the Every logo is not quite right." - He compares GPT 5.6's deck unfavorably: "a little bit worse... a little bit more bare... the arrows are not done right." ### Discernment / "McDonald's eval" - He describes a loosely named eval where models sycophantically agree with bad analogies, framing the issue as "people pleasing" or "discernment," and claims "Fable's level of discernment is much higher than previous models I've tried." - He says Fable 5.1 flagged a strategy question in a marketing meeting as worth his attention — "This is new." --- ## Writing Claims - **Historical claim**: "Anthropic has really struggled with writing recently... with Sonic [sic] 5 and Opus 5, pretty much everyone was just like, this model doesn't speak English correctly," and "everyone internally at Every had switched to ChatGPT." - **Current claim**: Fable 5.1 is "not only better than Opus at writing, it's actually starting to compete with ChatGPT," with "pretty clear prose" but "still a lot of the Claude literariness." - He shows a one-shot blog post from Mike Taylor's transcript as strong, non-slop prose ("It doesn't have any of the X, not Ys"), contrasting an Opus 5 output he calls "so dark" and "such a dick" (his characterization of the writing tone). - **Personal qualification**: He himself still uses GPT 5.6 ("in the chat GPT for work app") for most day-to-day writing because he prefers "minimal straight ahead pros" (sic) and finds 5.6 easier to collaborate with — but he now flips to Fable "when the first prompt or two to chat GPT doesn't work." - **Colleague attribution**: Katie Parrott, a staff writer, was "using chat GPT all the time, and now she's back to Claude for a lot of her writing." (Reported, not directly verified.) - **Editing example**: Working on a Codex history piece, he says the model correctly diagnosed why a paragraph failed ("goes flat, right where it should pay off") and "knows how sentences connect together." --- ## Usage-Pattern Claims (personal data) - Access began **August 24**. He claims his session count in Claude Code didn't change much but his **token usage jumped dramatically**, because he delegates "big, chunky" tasks and "lets it cook" while doing day-to-day interactive work in ChatGPT/Codex. - Token spend went to: Hands, a "personal feed," and "personal feed experiments" — big end-to-end builds. - Framing: "You want both gears" — Fable 5.1 "works at the cost and speed of Opus 5, but it gives you the delegation power of Fable that used to be pretty out of reach unless you were willing to spend a ton of money." --- ## Contradictions and Qualifications Preserved 1. **Enthusiasm vs. expense**: Calls the model cheap and token-efficient overall, but explicitly warns large UltraCode runs cost millions of tokens. 2. **Praises writing vs. keeps GPT 5.6**: Says Fable 5.1 "competes with ChatGPT" at writing, yet personally still uses ChatGPT for most daily writing. 3. **GPT 5.6 outperforms on one eval**: On survey analysis, he explicitly prefers GPT 5.6's narrative synthesis, despite the overall pro-Fable framing. 4. **Interactivity trade-off**: Fast on low-effort settings, but "basically just going to like go off and cook" on UltraCode — "a different usage pattern than you might be used to." 5. **Acknowledged imperfections**: "I have no idea how this works" (Hands); "Every logo is not quite right" (deck); prose is "still like a little bit literary and chunky." --- ## Notes on Source Fidelity - Several model names in the transcript appear garbled ("GBT 5.6 Sol," "Sonic 5," "Kieran Klassen" vs. later "Kieran Klaassen") and are preserved as spoken/where the transcript is the only evidence. - All quantitative figures (tokens, latency, dates, sub-agent counts) are as stated by the speaker and were not independently verified. - Third-party opinions (Klaassen, Parrott, Taylor) are recorded as reported speech only. # Speaker & Provenance Record — "We Tested Anthropic's Fable 5.1 for a Week" ## Speakers & Roles | Speaker | Role | Basis | |---|---|---| | **Dan Shipper** | Creator / host; self-identified co-founder and CEO of **Every** | Explicitly stated in transcript: *"My name is Dan Shipper, and I'm the co-founder and CEO of Every"* | No guests, narrators, or additional identified speakers appear. All speech is first-person by Shipper; team members are referenced but do not speak. ## Third-Party People Referenced (not speaking) - **Kieran Klaassen** — GM of Quora (per transcript) at Every; described as liking the model despite previously disliking Opus 5 and Sonnet 5. - **Katie Parrott** — staff writer at Every; described as returning to Claude for writing. - **Mike Taylor** — head of evals at Every; a transcript of his talking was used as input for a model-generated blog post. - **Sam Parr** — referenced as asking a question on X (quoted within model-generated output, not his own statement in this video). ## Claimed Sources & Evidence Channels - **Internal agent benchmark** at Every: claims of ~766 tokens per run vs. ~2,000 for Opus 5; ~22s latency vs. ~37s. - **Internal eval viewer** showing prompts and outputs for knowledge-work tests (data analysis, PowerPoint generation). - **Customer survey** (conducted "a couple years ago") fed to the model for data-analysis testing. - **Personal usage data**: access since **August 24**; token/usage graphs comparing Claude Code vs. ChatGPT/Codex. - **Reading-ease / grade-level scores** claimed for Fable 5.1 vs. Opus 5 and GPT 5.6 Sol. - **Model-generated artifacts shown**: "Hands" computer-use app, NPS dashboard, slide deck, blog post draft. > Note: Products referenced (Fable 5.1, Opus 5, GPT 5.6, Claude, ChatGPT, Codex, "McDonald's eval") are claims within the transcript; no external citations are provided. ## Promotional Segments (flagged) - **Every subscription promotion**: Shipper describes Every as *"the only subscription you need to stay at the edge of AI."* - **Closing promotion**: long-form written "vibe check" at **every.to** and a downloadable **prompt library** (directed to "Every.TL" in transcript). ## Comment Availability - No information about comments is present in the transcript or metadata. ## Uncertainty Notes - Transcript contains transcription artifacts (e.g., "Every.TL" vs. "every.to"; "Sonic 5" vs. "Sonnet 5"; "GBT 5.6"); recorded as heard. - Benchmarks and comparisons are self-reported by the creator with no independent verification in the source. It's model release day! We've got an all-new Fable, Fable 5.1, and it is Fable for everyone. We've been testing it for about a week. This is a sick model. It is a better coding model than the original Fable, so if you love the original Fable, you're going to love this model even more. And it speaks English. It actually talks to you like a normal person. It says things that you can understand, and that's not all. It is significantly faster. It's like twice as fast as old Fable, and it uses half the tokens as Opus 5 for similar tasks. So it's priced like a Fable, but you'll be able to use it for all your Opus tasks, and there's really no reason for you to use Opus anymore. Fable all day, baby. So let's get into the vibe check. So before we get into this video, who am I, and how do I possibly have a full review of a model that just dropped? My name is Dan Shipper, and I'm the co-founder and CEO of Every. Every is the only subscription you need to stay at the edge of AI. We get all of the models before they come out, and we test them with a team of writers, designers, engineers, and more to tell you what they're good for. So on the day a model drops, we have long-form reviews to tell you what to use them for and what not to use them for from our real experience doing real work. So we've been testing this model on all the kinds of things that you do every day. So we've done it from on coding, we've done it on knowledge work like decoration and data analysis, and we've done it on writing. And I'm going to show you everything that we found in our usage and how that has pulled through into how I use it day to day. What are the actual numbers for how it has changed my usage over the week since I got access to it? Let's get started. One of the coolest things I had to build is this thing called Hands. It is a computer use agent that sits on my computer and allows other agents to use my computer remotely. So that's the arrow for Hands. That's Hands going and using my computer. And you can see it has this little window on my computer that shows me what Hands is doing. And what this does is it lets me remotely use my computer from any agent I want. And this is like a hard thing to build. It's hard to make a Mac app that is a computer use agent that is accessible from the web that can do any task you ask it to do. And I tried to build Hands in 5.6, in regular Fable. And it is like good, but like it didn't really work. And this actually just works. Like it just uses my computer and I can I can have it use my computer from Slack. And that is like fucking wild. Look at that. It's look at that. Like that is fucking wild. And look, it's going to start typing too. And you can see it's working in ChatGPT. You can see it's typing. It's typing. And I have no idea how this works. This was built end to end by Fable 5.1 from a couple prompts. That is a desktop agent built fully by this model. So this is just like a fully fledged app. It has a screenshot of the current state. It says what it's doing. It has a log. Look, look, these are all the tasks I've been doing with with this thing. I mean, this would ordinarily take months to be good. And it just worked after like a day. I gave a prompt to Fable 5.1 and it just churned on UltraCode with like 40 sub agents for like a day. And then this came out and it just it just works. I don't know. I don't know how else to say it. It's a completely different experience of programming to be able to give a simple prompt to a model and have a completely usable, complex application come out like a day later. That said, there are some things to be aware of. It's more token efficient than Opus 5. But it's, you know, if you give it a big task like this on UltraCode, it's going to run for like a day and it's going to use I mean, this is like a three to five million token run. So it can get it can get expensive. And especially on UltraCode, it's not really that conversational. It is conversational on lower lower effort settings. But if you put it on UltraCode, it's basically just going to like go off and cook and come back later. So that's it's just like a different usage pattern than you might be used to. It's like better for especially for big projects, like not as interactive coding work, but fucking sick. And and the other technical people at every like Kieran Klassen, who's a GM of Quora and makes compound engineering love this model. Like I think for Kieran, he's been a Claude stan for a long time. And even he didn't like Opus 5, didn't like Sonnet 5. And this model, he's like, this is my favorite again. So it's like a big jump in both power and usability, which usually don't go hand in hand and speed. It's just fast enough to use for day to day tasks to give you a little bit of a sense for its level of efficiency and how and how fast it is. We ran it on our internal agent benchmark. This measures it on tasks that our internal agent does like work inside of a company. And we compared it to Opus 5. So in terms of the number of tokens per run, on average, this model consumed about 766 tokens for each run. So each request to the model is about 766 tokens. By comparison, Opus 5 was almost 2,000 tokens for the same tasks. Similarly, the average latency for this model was about 22 seconds. So that's how long it took for a basic request to get a response. 22 seconds. Opus was about 37 seconds. So it's about twice as fast as Opus. And it uses about half the tokens. That's really crazy. And it's a very rare jump. There was this moment in like December 2025, where Opus 4.5 and where Opus 4.5 and GPT 5.3-ish, like cross this barrier where suddenly all these vibe coding use cases opened up. And all these people discovered that you could code in this new way with these models. I think there's a similar thing happening with Fable 5.1 because it opens up these long horizon tasks in an easy to use form factor that's cheap enough for regular people to try. I think it has the chance to open up real delegation of coding and knowledge work tasks to two more average knowledge workers. And that's a big deal. So now let's get into knowledge work. So we have this really cool eval viewer. So I can show you exactly what we ran it on, what the prompts were, and what the outputs were. The first knowledge task we always do with this model is we run it on data analysis. So we feed it a survey that we gave, like a real survey that we gave to our customers a couple years ago, and we ask it to analyze the survey. So the prompt was, I want you to build a static HTML NPS dashboard. NPS is net promoter score and pick out interesting insights after doing the analysis. So this is what we got out of the model. So it's a very clean design. It really, it makes it really easy to see like the score distribution and all that kind of stuff, which is, I would say like pretty normal. You know, it's not, it's not obviously like AIE right now. So it doesn't feel like slap. One thing that it, that it did that I think is a little bit worse than for example, GPT 5.6 is, and we can actually go look at GPT 5.6 for comparison, like GPT 5.6. This is GPT 5.6's version. It pulled out a story at the top, a strong core, a clear promise, customer to see every is a rare mix of thoughtful. This is something that Fable 5.1 didn't really do. It was much more like straight ahead. So I, you know, so I really, I actually really like GPT 5.6 on this task. But one thing that I want to point out from 5.1 that's really interesting is it did an incredibly good job at pulling out real insights from both the quantitative and qualitative data and then presenting it to us. So like what people love, the writing is the core of the love, the apps are the bonus. Keeps me current without overwhelming me is the strongest value proposition. Interesting is a warning word, useful, great, and smart predict promoter. So this is really cool, right? So it's, it went and did the analysis to understand that when someone, one of our customers calls us interesting, they actually don't really like us that much. But if someone calls us useful, great, or smart, they do love us. And then it was able to package it, not only find the insight, but package it into a little sentence that could express it to me in a way that I didn't feel like was slop. This is something that Old Fable wouldn't do. Opus 5 absolutely would not do. And generally models struggle with. So there's a, there's a leap here in how it is able to pick out interesting things and then present them to you. Another thing that we always do is we have these models create a PowerPoint. So here's the PowerPoint it created. You know, you'll notice some like small visual things like this. Every logo is not quite right. But in general, the prompt we give it is just take compounding engineering and turn it into a deck. It created this deck end to end. I mean, this is a very well put together deck on a couple of different levels. For one, it does a very good job of pulling out specific ideas and making sure each slide has one idea on it. It does a really good job of, you know, these subtle visual design things like, you know, the yellow for the compound engineering and, and, and sort of the end of that paragraph. And it even does things like you'll, you'll see it placed these arrows to show the compound engineering loop. It created, it does these bubbles with these arrows. Most models can't get these arrows, right? There, there are these like subtle little things that it does really, really well. So it is the kind of model where if you give it a, you know, a well-written doc and turn it and ask it to turn it into a deck, it'll do a really good first pass end to end without very much interference, which is something that most models can't do. So for example, this is what 5.6's deck looks like. And you can tell it's just, it's like a little bit worse. It's a little bit more bare. You can see the arrows here. The arrows are not done right. So, so the thing I want you to take away is developers for at least the last six months, depending on who you are, have been getting used to delegating big chunks of work to the model. That hasn't really been possible for knowledge workers. You can do it, but the details aren't really right. It looks a little bit like slop. This is a big step in the direction of, actually, I can have it make a deck for me end to end. Actually, I can have it do a bunch of data analysis and get something back. And I'm like, this is actually pretty good. And to be clear, the original Fable did some of this, but not at this speed and not at this cost. So it makes it available to more and more people that previously would have looked at Fable and been like, this is not for me. For example, this is a meeting where I can be a tiebreaker. Our head of marketing is talking about a new launch and who we're targeting for that launch and whether we might want to target other customers as well. And it's flagging me as like, Dan, there's a question about strategy here. You might want to be involved in this. You might want to know about this. And this is like a superpower for me that Fable 5.1 can do is pick out interesting things that you're going to care about from what's happening in your company, what's happening in the world, and tell you about it in a way that kind of clicks for you. This is new. Normally it would show you stuff and you'd be like, ah, like this doesn't really click for me. It's like, it's not actually interesting. This is the first model that I've ever found that can pull out something interesting and then present it to you in a way that makes it click. I think the thing that Fable can do, well, one, it just like built this app totally end to end. But the thing that it can do is if when Fable thinks something is interesting, it's more likely to actually be interesting. You know, there's this, um, there's this eval that's like loosely called the McDonald's eval, where if you're talking to your model and you're like, oh, that's exactly like McDonald's. And it's like, oh yeah, you're totally right. But you're talking about something completely random. That's a, that's an example of the kind of thing that models struggle with is I would call it, I mean, to one, to some degree, it's like people pleasing to another. It's basically discernment. Like when you make a connection, is that connection actually interesting? Is it a real connection or is it just that you can connect anything to anything, which these models can do because they know everything. And I think Fable's level of discernment is much higher than previous models I've tried. And that takes me to the next one, which is writing. And obviously like writing is, that's, that's what I love. So Anthropic has really struggled with writing recently. They used to be the top writing model, but with Sonic 5 and Opus 5, pretty much everyone was just like, this model doesn't speak English correctly. It doesn't give me good output anymore. And everyone internally at every had switched to ChatGPT. With Fable 5.1, it's not only better than Opus at writing, it's actually starting to compete with ChatGPT. It writes quickly. It has pretty clear prose. It still has a lot of the Claude literariness. But if you look at the reading level that it writes at and the reading ease scores, it has the highest reading ease and the lowest grade level of Opus 5 and GBT 5.6 Sol. So it is just producing stuff that's easier to read. So as an example of the kind of one shot writing this thing does, we asked it to create a blog post based on a simple prompt and a long transcript of Mike Taylor, who's our head of evals, just talking about some of the work that he does. And we asked it to write some of the things that he's learned about getting teams to use AI. And this is the post that it created just from a very, very simple prompt. You can see the prompt is really small. Raise the ceiling, not the floor. Seven things I've learned about getting teams to actually use AI for my first few months running tech consulting at Every. Sam Parr asked me a question on an X last week that I've been asked in one form or another in almost every client meeting since I joined Every. How's everyone getting team adoption for Claude? It's the right question. The gap between what the terminally online people and a company can do with AI and what everyone else does with it has never been wider. This is like pretty good prose. It doesn't feel like slop. Buy the model. Buy the model direct, not third-party tools. When a company decides to do AI, the first instinct is to evaluate tools. Someone assembles a spreadsheet of vendors, the Clutter Codex or Gemini under the hood. It doesn't have any of the X, not Ys. It has the raise the ceiling, not the floor, but still it's not, it doesn't have that slop feeling. It's well-structured. It's thoughtful. It's actually really impressive. Let's look back at Opus, Opus 5. Same prompt, same exercise, same everything. Nobody wants to use your AI tools. Opus 5 was such a dick. Nobody wants to use your AI tools. Some of the things I've learned about AI adoption since joining every consulting. Sam Parr asked a question last week that I think is the real question in AI right now. So that opener, it's a sort of similar opener. It's a little bit of more of a minimal sentence construction, which isn't necessarily bad, but you can just sort of feel, let's see. The instinct is to go after laggards. We bought you the tools. Now everyone needs to use AI. This does not work. On pain of death, a good number of people are emotionally unwilling to use AI. Like, what is this thing even talking about? It's, Opus 5 is just so dark and they fixed it with Fable 5.1. It's really impressive. I find for me personally, I'm, I'm a pretty, I write every day and I'm still using 5.6 in the chat GPT for work app for most of my day-to-day writing. I still find with Fable 5.1, it's still like a little bit literary and chunky, but different writers have different styles. So Katie Parrott, who's a staff writer uses, is now back. She, she was using chat GPT all the time, and now she's back to Claude for a lot of her writing because she's a big Claude fan for many years until the last few models. So I find that for me, I really like more minimal straight ahead pros and I can kind of collaborate with 5.6 a little bit easier to, to create that. But I am now flipping back to 5.1 to Fable when the first prompt or two to chat GPT doesn't work. So I find it to be a valuable part of my arsenal where I wasn't using Claude at all for writing anymore. And that's new. And I certainly, I mean, I certainly was never using Fable for writing because Fable is like this warp drive thing. So one of the things this model knows how to do that other models don't really know is it knows how sentences connect together. Like I was working on this piece. I'm working on this history of Codex. And I was working on a piece of it where I asked a question, why was Codex Cloud different? And the details of this don't matter. Why was Codex Cloud different? And then there was some text after that. There was, I had it, I had a, you know, I had a following paragraph and I just knew it wasn't really quite working. And I threw it into this model and it was like, here's what's dragging it down. Roughly in order of how much it matters. The biggest problem is that the paragraph goes flat, right where it should pay off. You set up a real puzzle. Why was Codex Cloud different? And then answer it in the voice of a help center article, the right software dependencies, blah, blah, blah. Everywhere else in the piece you reach for an image. Here you explain dependency hell literally. So it's actually zeroing in on the right part of the problem and then telling me how to fix it. And the takes that it does, generally the sentence are connecting there. It's leading from one sentence to the next. It doesn't write as many of those sentences where you're like, oh yeah, that's, that sounds right. And then you think about it and you're like, that doesn't mean anything. I have no idea what it means. And that's new for a model of this level of power. So the takeaway here is if you really liked Claude models until Opus 5 or Sonic 5 for writing, you're going to really like this model. It's going to be back to the stuff that you love. And if you've been using GPT 5.6 or similar models for your writing, you should give this one a try. It's a good one-two punch to give you a little bit more variation in perspective in your writing. That'll help liven it up. So where does this leave us? I want to show you like my actual usage. So I got this model on August 24th and you can see, you know, this is in, in green, this is Codex or ChatGPT for work and in orange, it's Claude code. You can see it. My, my level of use has not really changed. I'm mostly using Codex, but then if I look at the number of tokens, my, my use of Claude jumped dramatically. Why is that? That's interesting, right? What Fable 5.1 has done for me is given me a model that I set off on a big, chunky task and I just let it cook. And I'm in ChatGPT for my more basic day-to-day stuff. And then I flip back to Fable every once in a while. And so I'm getting like all this work done, but I'm not really checking it because it, it just does the work end to end. And this is that new style of coding and knowledge work that I'm talking about. You want both, you want both gears, but what, what Fable 5.1 has really given me is this extra tool that works at the cost and speed of Opus 5, but it gives you the delegation power of Fable that used to be pretty out of reach unless you were willing to spend a ton of money. So you can sort of see that in the graph, right? Like I'm still using a lot of ChatGPT tokens, but I'm using way more Claude tokens and I'm in ChatGPT more. I'm, I'm prompting ChatGPT more. It's more of an interactive model. I think that's really interesting. Okay. What did I actually use the tokens on? It's like it's hands. It's this personal feed experiments. It's these big things that I'm building end to end. And the rest of my work, I'm mostly still doing in ChatGPT. So final thoughts, if you like Claude models and you've been disappointed by the recent releases, you're going to love this. You need to check it out. And if you're a big GPT 5.6 fan, like I am, it's worth really adding to your arsenal, especially for your big projects and especially for trying to understand, okay, where's knowledge work going? How is knowledge work going to work when I'm delegating lots of things like making a slide deck end to end that used to be impossible. It's starting to be possible now. And so if you want to get a glimpse of this new way of working in a way that isn't going to totally break your bank account, check out Fable 5.1. In addition to this video, we have a long form written vibe check with all of our benchmarks and all of our testing live now on every.to. We also have a prompt library that you can download to help you get the most out of Fable 5.1. So if you want to go deeper after this video, Check it out, Every.TL.