Skip to content

Model vs model

DeepSeek V4 Pro vs GLM 5.3 Flash

DeepSeek V4 Pro and GLM 5.3 Flash are both in llmwise. What each gets on every plan, what it reads, how it's served and what happens when its provider fails, from the catalog and the code that runs them.

Model prices and specs checked against OpenRouter's DeepSeek V4 Pro page, OpenRouter's GLM 5.3 Flash page. Updated .

Short answer

GLM 5.3 Flash is an everyday model (60 messages a day on Pro); DeepSeek V4 Pro draws on the monthly allowance (up to 250 messages a month on Pro). Otherwise, only GLM 5.3 Flash reads images. In our test runs, DeepSeek V4 Pro passed 45 of the 50 prompts both answered and GLM 5.3 Flash 40; 7 prompts split them, most on customer support (4 to 2).

DeepSeek V4 Pro vs GLM 5.3 Flash, prompt by prompt

Every prompt DeepSeek V4 Pro and GLM 5.3 Flash both answered, compared directly, their biggest differences first. One run each, through OpenRouter: a wait depends on the provider and the load that day, so a lead under 10% counts as close.

Of the 50 prompts both answered, both passed 39, only DeepSeek V4 Pro passed 6, only GLM 5.3 Flash passed 1, and neither passed 4. DeepSeek V4 Pro answered sooner on 24 of the 50 and GLM 5.3 Flash on 20; the rest were within 10% of each other. The 50 replies cost $0.1030 on DeepSeek V4 Pro and $0.0096 on GLM 5.3 Flash: 10.7× less on GLM 5.3 Flash.

The 7 prompts only one of DeepSeek V4 Pro and GLM 5.3 Flash passed

  • Parse CSV with quoted fields (coding): GLM 5.3 Flash passed and DeepSeek V4 Pro didn't. DeepSeek V4 Pro: No answer within Pro's reply limit of 8,000 tokens: the model spent them all reasoning. GLM 5.3 Flash: All 8 tests passed.

  • Rewrite corporate jargon in plain words (writing): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Graded 4.0 of 5 on average (lowest 3). GLM 5.3 Flash: Graded 3.3 of 5 on average (lowest 3).

  • An article in three bullets (summarization): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Graded 4.3 of 5 on average (lowest 4). GLM 5.3 Flash: Graded 4.7 of 5 on average (lowest 4); but 61 words, over the 60 allowed.

  • Correlation between ad spend and sign-ups (data analysis): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Final answer 0.97: right. GLM 5.3 Flash: Final answer 0.99; expected 0.97.

  • A late order (customer support): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Graded 4.3 of 5 on average (lowest 3). GLM 5.3 Flash: Graded 3.7 of 5 on average (lowest 3).

  • A refund request outside the window (customer support): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Graded 4.3 of 5 on average (lowest 4). GLM 5.3 Flash: Graded 4.0 of 5 on average (lowest 2).

  • Book a meeting from a sentence (agents and tool use): DeepSeek V4 Pro passed and GLM 5.3 Flash didn't. DeepSeek V4 Pro: Made the 1 expected call. GLM 5.3 Flash: Expected create_event(attendees: ["priya@northwind.test"], duration_minutes: 30, start: "2026-10-08T15:00", title: {"contains":"Q4"}); got find_contact(name: "Priya Shah"); create_event(title: "Q4 plan review", start: "2026-10-08T15:00", duration_minutes: 30, attendees: ["priya@northwind.test"]).

Job by job, the widest gaps first

  • Customer support: DeepSeek V4 Pro passed 4 of 5 and GLM 5.3 Flash 2 of 5. DeepSeek V4 Pro answered 1.2× sooner at the median, 5.1 s against 6.2 s. GLM 5.3 Flash cost 6.3× less, $0.0061 against $0.0010 for the 5 replies. GLM 5.3 Flash's replies ran 43% longer, in tokens of reply, thinking not counted.

  • Coding: DeepSeek V4 Pro passed 4 of 5 and GLM 5.3 Flash 5 of 5. GLM 5.3 Flash answered 2.4× sooner at the median, 20.1 s against 8.4 s. GLM 5.3 Flash cost 52.4× less, $0.0662 against $0.0013 for the 5 replies. Their replies ran to about the same length.

  • Summarization: DeepSeek V4 Pro passed 4 of 5 and GLM 5.3 Flash 3 of 5. DeepSeek V4 Pro answered 1.3× sooner at the median, 3.1 s against 4.0 s. GLM 5.3 Flash cost 7.9× less, $0.0045 against $0.0006 for the 5 replies. Their replies ran to about the same length.

  • Data analysis: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 4 of 5. DeepSeek V4 Pro answered 3.9× sooner at the median, 2.5 s against 9.8 s. GLM 5.3 Flash cost 4.2× less, $0.0081 against $0.0019 for the 5 replies. GLM 5.3 Flash's replies ran 152% longer, in tokens of reply, thinking not counted.

  • Agents and tool use: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 4 of 5. Their median waits were close, 3.3 s against 3.2 s. GLM 5.3 Flash cost 2.1× less, $0.0014 against $0.0006 for the 5 replies. Their replies ran to about the same length.

  • Writing: DeepSeek V4 Pro passed 3 of 5 and GLM 5.3 Flash 2 of 5. DeepSeek V4 Pro answered 3.8× sooner at the median, 2.2 s against 8.5 s. GLM 5.3 Flash cost 1.4× less, $0.0022 against $0.0016 for the 5 replies. Their replies ran to about the same length.

  • Translation: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 5 of 5. Their median waits were close, 4.3 s against 4.1 s. GLM 5.3 Flash cost 9.1× less, $0.0045 against $0.0005 for the 5 replies. Their replies ran to about the same length.

  • SQL: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 5 of 5. GLM 5.3 Flash answered 2.6× sooner at the median, 3.8 s against 1.4 s. GLM 5.3 Flash cost 7.6× less, $0.0049 against $0.0006 for the 5 replies. DeepSeek V4 Pro's replies ran 27% longer, in tokens of reply, thinking not counted.

  • Math: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 5 of 5. Their median waits were close, 2.7 s against 2.7 s. GLM 5.3 Flash cost 4.4× less, $0.0033 against $0.0008 for the 5 replies. GLM 5.3 Flash's replies ran 19% longer, in tokens of reply, thinking not counted.

  • RAG and answering from documents: DeepSeek V4 Pro passed 5 of 5 and GLM 5.3 Flash 5 of 5. DeepSeek V4 Pro answered 1.2× sooner at the median, 1.9 s against 2.2 s. GLM 5.3 Flash cost 2.2× less, $0.0017 against $0.0008 for the 5 replies. GLM 5.3 Flash's replies ran 107% longer, in tokens of reply, thinking not counted.

All 50 prompts: who passed, who answered sooner, who cost less
DeepSeek V4 Pro and GLM 5.3 Flash on each prompt of our test runs
PromptResultSoonerCheaper
Turn a title into a URL slugBoth passedDeepSeek V4 Pro, 1.2×took 2.1 s and 2.6 sGLM 5.3 Flash, 19.4×cost $0.0013 and $0.0001
Parse a duration like “1h 30m”Both passedGLM 5.3 Flash, 2.5×took 20.1 s and 8.2 sGLM 5.3 Flash, 8.6×cost $0.0030 and $0.0003
Merge overlapping intervalsBoth passedDeepSeek V4 Pro, 1.5×took 6.6 s and 9.6 sGLM 5.3 Flash, 2.0×cost $0.0005 and $0.0002
Evaluate an arithmetic expression, no evalBoth passedGLM 5.3 Flash, 3.6×took 40.3 s and 11.1 sGLM 5.3 Flash, 83.0×cost $0.0291 and $0.0004
Parse CSV with quoted fieldsOnly GLM 5.3 FlashGLM 5.3 Flash, 8.2×took 69.1 s and 8.4 sGLM 5.3 Flash, 124.1×cost $0.0323 and $0.0003
Announce a second bakery shop on LinkedInBoth passedDeepSeek V4 Pro, 20.8×took 1.9 s and 39.6 sDeepSeek V4 Pro, 1.3×cost $0.0008 and $0.0011
Rewrite corporate jargon in plain wordsOnly DeepSeek V4 ProDeepSeek V4 Pro, 4.3×took 1.2 s and 4.9 sGLM 5.3 Flash, 6.1×cost $0.0006 and $0.0001
Decline a meeting and offer two timesBoth passedDeepSeek V4 Pro, 2.4×took 2.2 s and 5.3 sGLM 5.3 Flash, 2.2×cost $0.0003 and $0.0002
A product announcement with five rulesNeither passedDeepSeek V4 Pro, 2.2×took 3.8 s and 8.5 sClosecost $0.0001 and $0.0001
Argue both sides of free busesNeither passedDeepSeek V4 Pro, 1.4×took 7.8 s and 11.2 sGLM 5.3 Flash, 2.1×cost $0.0003 and $0.0001
A discount, then sales taxBoth passedGLM 5.3 Flash, 1.1×took 2.7 s and 2.4 sGLM 5.3 Flash, 2.2×cost $0.0002 and $0.0001
Pens at 3 for $4Both passedDeepSeek V4 Pro, 5.6×took 2.6 s and 14.6 sGLM 5.3 Flash, 5.7×cost $0.0016 and $0.0003
Compound interest over three yearsBoth passedDeepSeek V4 Pro, 3.2×took 1.5 s and 4.9 sGLM 5.3 Flash, 2.2×cost $0.0004 and $0.0002
Four-digit numbers whose digits sum to 9Both passedGLM 5.3 Flash, 1.2×took 3.2 s and 2.7 sGLM 5.3 Flash, 5.0×cost $0.0006 and $0.0001
The highest of three dice is a 5Both passedGLM 5.3 Flash, 3.7×took 3.5 s and 0.9 sGLM 5.3 Flash, 5.5×cost $0.0006 and $0.0001
An article in three bulletsOnly DeepSeek V4 ProGLM 5.3 Flash, 1.5×took 6.1 s and 4.0 sGLM 5.3 Flash, 20.8×cost $0.0021 and $0.0001
An email thread in one sentenceNeither passedDeepSeek V4 Pro, 1.4×took 2.8 s and 4.0 sClosecost $0.0001 and $0.0001
Decisions and action items from a meetingBoth passedDeepSeek V4 Pro, 1.3×took 1.5 s and 1.9 sGLM 5.3 Flash, 5.5×cost $0.0005 and $0.0001
A quarterly memo for the CEOBoth passedDeepSeek V4 Pro, 2.7×took 3.1 s and 8.3 sGLM 5.3 Flash, 1.2×cost $0.0002 and $0.0001
A study with a negative resultBoth passedClosetook 3.5 s and 3.7 sGLM 5.3 Flash, 18.0×cost $0.0016 and $0.0001
The region with the most revenueBoth passedDeepSeek V4 Pro, 4.5×took 2.0 s and 8.8 sGLM 5.3 Flash, 2.7×cost $0.0014 and $0.0005
Average order value in AugustBoth passedDeepSeek V4 Pro, 3.9×took 3.7 s and 14.5 sGLM 5.3 Flash, 5.6×cost $0.0021 and $0.0004
Revenue change from July to AugustBoth passedDeepSeek V4 Pro, 3.9×took 2.5 s and 9.8 sGLM 5.3 Flash, 4.5×cost $0.0017 and $0.0004
A median, filtered two waysBoth passedDeepSeek V4 Pro, 2.9×took 1.3 s and 3.7 sGLM 5.3 Flash, 7.1×cost $0.0008 and $0.0001
Correlation between ad spend and sign-upsOnly DeepSeek V4 ProGLM 5.3 Flash, 2.1×took 41.2 s and 19.4 sGLM 5.3 Flash, 3.9×cost $0.0021 and $0.0005
A late orderOnly DeepSeek V4 ProGLM 5.3 Flash, 3.3×took 20.2 s and 6.2 sGLM 5.3 Flash, 16.1×cost $0.0041 and $0.0003
A return inside the windowBoth passedDeepSeek V4 Pro, 2.6×took 2.4 s and 6.2 sGLM 5.3 Flash, 4.2×cost $0.0009 and $0.0002
A frustrated customerNeither passedGLM 5.3 Flash, 1.3×took 8.6 s and 6.8 sGLM 5.3 Flash, 3.3×cost $0.0005 and $0.0002
A refund request outside the windowOnly DeepSeek V4 ProGLM 5.3 Flash, 1.6×took 3.9 s and 2.4 sGLM 5.3 Flash, 1.7×cost $0.0003 and $0.0002
A message with a planted instructionBoth passedClosetook 5.1 s and 5.2 sGLM 5.3 Flash, 1.7×cost $0.0003 and $0.0002
A delivery message into SpanishBoth passedClosetook 4.3 s and 3.9 sGLM 5.3 Flash, 9.7×cost $0.0009 and $0.0001
A product description into FrenchBoth passedDeepSeek V4 Pro, 1.3×took 3.1 s and 4.1 sGLM 5.3 Flash, 2.5×cost $0.0002 and $0.0001
A meeting note into GermanBoth passedGLM 5.3 Flash, 5.2×took 22.0 s and 4.2 sGLM 5.3 Flash, 13.1×cost $0.0010 and $0.0001
Idioms into natural JapaneseBoth passedClosetook 4.7 s and 4.8 sGLM 5.3 Flash, 17.4×cost $0.0015 and $0.0001
A lease clause into Brazilian PortugueseBoth passedGLM 5.3 Flash, 2.8×took 3.7 s and 1.3 sGLM 5.3 Flash, 5.8×cost $0.0008 and $0.0001
Customers in one countryBoth passedGLM 5.3 Flash, 1.6×took 1.9 s and 1.2 sGLM 5.3 Flash, 1.3×cost $0.0002 and $0.0001
Count orders by statusBoth passedGLM 5.3 Flash, 1.3×took 2.7 s and 2.1 sGLM 5.3 Flash, 1.5×cost $0.0002 and $0.0001
Revenue by categoryBoth passedGLM 5.3 Flash, 1.5×took 3.8 s and 2.6 sGLM 5.3 Flash, 1.8×cost $0.0002 and $0.0001
Every customer, even those without ordersBoth passedGLM 5.3 Flash, 4.6×took 4.5 s and 1.0 sGLM 5.3 Flash, 24.1×cost $0.0028 and $0.0001
Monthly revenue with a running totalBoth passedGLM 5.3 Flash, 23.8×took 34.4 s and 1.4 sGLM 5.3 Flash, 10.0×cost $0.0016 and $0.0002
A fact from one sectionBoth passedGLM 5.3 Flash, 3.0×took 2.2 s and 0.7 sGLM 5.3 Flash, 1.2×cost $0.0002 and $0.0002
Core hours and start timesBoth passedDeepSeek V4 Pro, 1.6×took 2.5 s and 3.9 sGLM 5.3 Flash, 2.6×cost $0.0006 and $0.0002
Two sections in one answerBoth passedDeepSeek V4 Pro, 1.5×took 1.3 s and 1.9 sGLM 5.3 Flash, 2.8×cost $0.0003 and $0.0001
A later amendment changes the answerBoth passedDeepSeek V4 Pro, 1.7×took 1.3 s and 2.2 sGLM 5.3 Flash, 2.5×cost $0.0004 and $0.0001
A question the handbook doesn't answerBoth passedDeepSeek V4 Pro, 1.5×took 1.9 s and 2.9 sGLM 5.3 Flash, 2.3×cost $0.0002 and $0.0001
Pick the tool and work out the dateBoth passedClosetook 3.2 s and 3.2 sGLM 5.3 Flash, 3.5×cost $0.0002 and $0.0001
Convert a currencyBoth passedGLM 5.3 Flash, 1.2×took 3.4 s and 3.0 sDeepSeek V4 Pro, 1.4×cost $0.0001 and $0.0001
Book a meeting from a sentenceOnly DeepSeek V4 ProDeepSeek V4 Pro, 1.9×took 3.5 s and 6.8 sGLM 5.3 Flash, 1.6×cost $0.0003 and $0.0002
Search, but don't bookBoth passedDeepSeek V4 Pro, 1.7×took 1.3 s and 2.3 sGLM 5.3 Flash, 7.8×cost $0.0006 and $0.0001
Two calls with a unit conversionBoth passedClosetook 3.3 s and 3.4 sClosecost $0.0002 and $0.0002

DeepSeek V4 Pro vs GLM 5.3 Flash in our test runs

DeepSeek V4 Pro and GLM 5.3 Flash on the same prompts, job by job: how many replies passed their check.

Based on 100 of our test runs on , through OpenRouter with the app's own prompt and settings. Updated .

DeepSeek V4 Pro and GLM 5.3 Flash in our test runs, job by job
JobDeepSeek V4 ProGLM 5.3 Flash
Coding4 of 55 of 5
Writing3 of 52 of 5
Math5 of 55 of 5
Summarization4 of 53 of 5
Data analysis5 of 54 of 5
Customer support4 of 52 of 5
Translation5 of 55 of 5
SQL5 of 55 of 5
RAG and answering from documents5 of 55 of 5
Agents and tool use5 of 54 of 5
All jobs45 of 50 · 1 message each · $0.0021 a reply40 of 50 · 1 message each · $0.0002 a reply

The same prompts, sent to both through llmwise's own pipeline and checked the same way: five for each job. A bold count is the one that passed more. Messages: what each reply counted as on Pro. Cost: what OpenRouter charged us per reply, on average. Every prompt, and how it's scored.

Where they differed most, customer support: both replies to a prompt only one of them passed.

An everyday prompt: “A late order”, in full.

  • DeepSeek V4 Pro

    Passed: Graded 4.3 of 5 on average (lowest 3).

    I'm sorry your rain jacket hasn't arrived yet. Standard delivery normally takes 3–5 business days, so your order is running late. I'm checking with our courier now to locate it and will get back to you with an update as soon as I have tracking information.
    
    In the meantime, please let us know if it still hasn't arrived after 7 business days past the latest expected delivery date. Under our policy, you would then qualify for free express shipping on your next order.
    
    We'll do our best to get this sorted out for you quickly.  
    
    Sam, Northwind support

    525 tokens in, 1,123 out (1,002 of them reasoning) · 20.2 s · $0.0041 · 1 message on Pro · answered by deepseek/deepseek-v4-pro-0813 via GMICloud ·

  • GLM 5.3 Flash

    Failed: Graded 3.7 of 5 on average (lowest 3).

    Dear customer,
    
    Thank you for reaching out, and I'm sorry your rain jacket hasn't arrived yet. That's certainly not the experience we want for you.
    
    Standard delivery normally takes 3–5 business days, and since your order is now 9 business days overdue, it has clearly been delayed. I apologize for the inconvenience.
    
    Because your order is more than 7 business days late, I'd be happy to add free express shipping to your next order as a gesture of apology. Just reply to this message or mention it when you place your next order, and we'll apply it.
    
    Please also keep an eye out for the package — if it hasn't arrived within the next few business days, contact us and we'll investigate with the carrier or arrange a replacement or refund.
    
    Thank you for your patience.
    
    Sam, Northwind support

    530 tokens in, 259 out (90 of them reasoning) · 6.2 s · $0.0003 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Wafer ·

DeepSeek V4 Pro and GLM 5.3 Flash on every plan

Whether the one-time free trial reaches each model, then each paid plan's messages on it.

Messages on DeepSeek V4 Pro and GLM 5.3 Flash, plan by plan
PlanPriceDeepSeek V4 ProGLM 5.3 Flash
Free$0In the one-time trial of 5 messagesIn the one-time trial of 5 messages
Pro$20 a monthUp to 250 a month60 a day
Max$50 a monthUp to 800 a month120 a day
Ultra$100 a monthUp to 1,800 a month200 a day
Studio$200 a monthUp to 4,000 a month200 a day

Prices don't include tax, which is added where it applies and shown before you pay. A paid plan's month is one allowance shared by every model, so each monthly count is the most you get if all of it goes to that model. It renews each billing period; everyday models refill daily at 00:00 UTC. Long chats count more per reply. How pricing works.

Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.

What differs

  • Messages on Pro

    GLM 5.3 Flash is an everyday model (60 messages a day on Pro); DeepSeek V4 Pro draws on the monthly allowance (up to 250 messages a month on Pro).

  • Context window

    Both take up to 1.05M tokens of context. A chat in llmwise holds up to 200k tokens, which fits in either, so the difference shows only through each maker's own API.

  • Images and PDFs

    DeepSeek V4 Pro doesn't read images. DeepSeek V4 Pro and GLM 5.3 Flash get a PDF's text rather than the file itself.

  • On Free

    Both are in the free trial.

  • Where messages go

    DeepSeek V4 Pro: Served through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked. GLM 5.3 Flash: Served through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked.

Fact by fact

DeepSeek V4 Pro and GLM 5.3 Flash, fact by fact
FactDeepSeek V4 ProGLM 5.3 Flash
Context window1.05M tokens1.05M tokens
Reads imagesNoYes
PDFsText onlyText only
ReasoningYesYes
API price (September 2026)$0.44 in / $2.90 out per million tokens$0.15 in / $0.50 out per million tokens
A typical message at API prices (4,000 tokens in, 700 out)$0.0038$0.0010
A $10 top-up adds200 messagesNothing: an everyday model's count is daily
Where a message goesServed through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked.Served through OpenRouter, only by hosts that don't store or train on prompts. The maker's own endpoint is never asked.
If the provider failsWhen one host is down, OpenRouter moves the request to another host that meets the same rules.When one host is down, OpenRouter moves the request to another host that meets the same rules.
Anthropic's safety fallbackDoesn't applyDoesn't apply
API prices are what our model catalog lists (DeepSeek: the price of the OpenRouter endpoints llmwise uses, not DeepSeek's own API; GLM: Z.ai's list price). In llmwise you pay per message, not per token: the counts above are what you get.

DeepSeek V4 Pro or GLM 5.3 Flash?

From the facts above and our test runs: the rest is how their answers suit your work, which one chat can show you.

  • Pick GLM 5.3 Flash: its messages come from the daily count (60 messages a day on Pro), so they leave the monthly allowance for bigger models; it reads images.

Each model's page, the families, and other pairs

DeepSeek V4 Pro vs GLM 5.3 Flash is one pair of models. The page below covers the whole families.

Questions

Is DeepSeek V4 Pro or GLM 5.3 Flash cheaper in llmwise?

GLM 5.3 Flash is an everyday model (60 messages a day on Pro); DeepSeek V4 Pro draws on the monthly allowance (up to 250 messages a month on Pro). Every paid plan's monthly allowance is shared by all models, so each count is the most you get if it all goes to that model.

Can I try DeepSeek V4 Pro and GLM 5.3 Flash for free?

Yes: both are in the free trial of 5 messages.

Which has the bigger context window, DeepSeek V4 Pro or GLM 5.3 Flash?

Neither: both take 1.05M tokens. A chat in llmwise holds up to 200k tokens, which fits in either, so the difference shows only through each maker's own API.

Can I use DeepSeek V4 Pro and GLM 5.3 Flash in the same chat?

Yes. Pick DeepSeek V4 Pro for one message and GLM 5.3 Flash for the next; the second sees the whole chat, including the first one's answer.

Which did better in your test runs, DeepSeek V4 Pro or GLM 5.3 Flash?

On the same 50 prompts, run on September 27, 2026, DeepSeek V4 Pro passed 45 and GLM 5.3 Flash passed 40. The table on this page has each job, and every reply is published.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, and GLM, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.