cYpher.camp

Choosing an AI model

Which AI gets the job done for less?

Give three models a simple job. Then give them a harder one. See when a cheaper model works and when it pays to switch.

Compare the same job Free demo. No sign-up or credits needed.
An everyday hatchback and a Porsche parked outside a pizza shop.
Would you send a Porsche to deliver one pizza?

Which AI can finish
for the fewest credits?

Try the email first. Then give the same models a harder job.

Pick a sample job
Change models GPT-6 Luna · GPT-6.2 Sol · Claude Fable 5.2

Pick up to three. A fourth pick replaces the first.

Sample email
“Your piza arrives at 6:00.”
One passing edit
“Your pizza arrives at 6:00.”

Credits are how you pay for AI in Studio. This free demo estimates how many each job takes.

GPT-6 Luna

OpenAI · Quick tasks

Est. credits
0
Try
0
Est. work time
0 min

Ready

GPT-6.2 Sol

OpenAI · Everyday work

Est. credits
0
Try
0
Est. work time
0 min

Ready

Claude Fable 5.2

Anthropic · Hard problems

Est. credits
0
Try
0
Est. work time
0 min

Ready

These jobs use sample results. Credit estimates include every try at each model’s listed rate.

Help me pick an AI

What do you need done?

Tell us the job and what matters most. Get three models to start with.

Choose a job
What matters most?

Your shortlist · Fix or write code

Try one. Check it. Switch if you need to.

We suggest these models based on what their makers say they do well. Your plan sets which ones you can use. AI Studio and AI Hires share the credits in your plan. In Studio, you’ll see an estimate before you send.

See plans for this work

03 / Follow the work

Credits follow the work.
Not the clock.

You do not pay for waiting. You pay for what the model reads and writes. Tap each step to see where your credits go.

Read 8,000Written 0example tokens so far. More tokens, more credits.

Step 01

It reads everything you send

Your instructions, files and chat history all count. The more it has to read, the more credits it uses.

Illustration, not a model speed comparison.

04 / Compare every model

All 15 models.
Side by side.

Every dot is a model. Further right costs more. Tap up to three and compare them line by line.

One medium task. Credits for one example request: the model reads about 30,000 words and writes about 6,000.

Quick tasksLowest cost. Best for simple, clear jobs.
Everyday workThe balance of quality and cost for most work.
Hard problemsFor long, tricky work that needs more thinking.

The three lanes are our picks for where to start, not test scores. A higher lane does not always cost more.

One medium task: lowest to highest

Side by side

GLM 5.3 FlashClaude Fable 5.2
Made byZ.aiAnthropic
Good fit forBudget-friendly coding and analysis of text and images.The hardest technical problems and open-ended projects.
Our laneA starting point, not a test scoreQuick tasksHard problems
One medium taskExample credits2 credits52 credits
ReadingCredits per 1M tokens30640
WritingCredits per 1M tokens1003,200
Can take inText, Images, Video, FilesText, Images
ThinkingWorking a problem through before answeringAlways onAlways on
Open weightsAnyone can download and run the modelYesNot listed
ReleasedThe day its maker released itAug 26, 2026Nov 9, 2026
Next stepTry it Maker's factsTry it Maker's facts
See every model in one table
ModelMade byOur laneMedium taskReadingWritingCan take inThinkingOpen weightsReleased by maker
DeepSeek FlashDeepSeekQuick tasks560240Text, ImagesCan be switched offNot listedSep 10, 2026
Gemini 3.5 Flash-LiteGoogleQuick tasks760500Text, Images, Audio, Video, FilesAdjustableNot listedJul 21, 2026
GLM 5.3 FlashZ.aiQuick tasks230100Text, Images, Video, FilesAlways onYesAug 26, 2026
Claude Haiku 5.5AnthropicQuick tasks220100Text, ImagesAdjustableNot listedOct 7, 2026
Gemini 4 HydrogenGoogleEveryday work10120600Text, Images, Audio, Video, FilesAdjustableNot listedNov 12, 2026
Gemini 4 NeonGoogleHard problems202401,200Text, Images, Audio, Video, FilesAdjustableNot listedNov 18, 2026
Gemini 4 CarbonGoogleHard problems13160800Text, Images, Audio, Video, FilesAdjustableNot listedNov 10, 2026
GPT-6 LunaOpenAIQuick tasks220100Text, ImagesCan be switched offNot listedSep 22, 2026
GPT-6.1 TerraOpenAIEveryday work780400Text, ImagesCan be switched offNot listedOct 26, 2026
GPT-6.2 SolOpenAIEveryday work13160800Text, ImagesAlways onNot listedNov 17, 2026
GPT-6.2 AstraOpenAIHard problems526403,200Text, ImagesAlways onNot listedNov 17, 2026
Kimi K3.1 ProMoonshotHard problems324002,000Text, Images, VideoAlways onAnnouncedOct 24, 2026
Claude Sonnet 5.5AnthropicEveryday work13160800Text, ImagesAdjustableNot listedSep 28, 2026
Claude Opus 5.5AnthropicHard problems263201,600Text, ImagesAlways onNot listedSep 22, 2026
Claude Fable 5.2AnthropicHard problems526403,200Text, ImagesAlways onNot listedNov 9, 2026

One thing they share: each maker lists a context window of about 1 million tokens, so how much a model can hold at once is not what sets these apart. Credit numbers are examples at today’s rates for short requests, with no tools and no saved-context discounts. Release dates are the day each maker released the model, not the day it arrived in cYpher.camp. Those dates and the facts about inputs, thinking and open weights come from each maker’s own pages, checked November 17–21, 2026. “Not listed” means the maker’s page does not say. What you can use in AI Studio may differ.

05 / Meet the whole lineup

No model wins
every job.

Explore all 15 by what they can help you do.

Z.ai

GLM 5.3 Flash

Budget-friendly coding and analysis of text and images.

A starting point, based on what its maker says this model is good at. Your instructions, tools and thinking settings change the result too.

Try this model Read the provider source ↗

What the makers’ own tests show

Higher price per token.
Sometimes a lower bill.

~½

Half the cost on one coding test.

Anthropic reports that Fable 5.2 at High effort matched Opus 5.5’s best FrontierCode result for about half the cost per task.

Read the evaluation ↗
≠

Writing faster is not finishing faster.

In Anthropic’s speed test, Sonnet finished before Haiku. Haiku wrote faster, but Sonnet needed fewer tokens to get there.

Read the experiment ↗

These are the makers’ own results under their own test conditions. They are not cYpher tests and they do not predict your job. Picks last reviewed November 12, 2026.

Your next step

Try it on your own work.

Give a model one real task in AI Studio. Check its answer. When the job changes, compare your options again.

Tried a model? Found something useful? Compare notes with people in Camp →