Side by side
| Fast | Smart | Premium models | |
|---|---|---|---|
| Best for | Edits, searches, summaries, routine fixes | Features across many files, hard bugs, reviews, plans | Work that needs the strongest Claude or GPT model |
| Score on the Artificial Analysis Intelligence Index | 42 | 46 | Claude Opus 5.5: 58 |
| Writing speed | 66 tokens a second | 47 tokens a second | Depends on the model |
| First text after | 2.7 s | 4.1 s | Depends on the model |
| Our test, 8 agent tasks | 5 done, 3 done in part, in 28 minutes | 8 done, in 116 minutes | Not in the test |
| Paid from | Uno AI hours | Uno AI hours | A monthly credit or a top-up, per token |
| On Pro+AI, after 45 full-speed hours | Keeps going at standard speed | Only within the full-speed hours | Credit, then top-up |
Minutes to work through the same agent tasks
Our test: 8 coding tasks, run by the same agent with each model. Fewer minutes means fewer AI hours spent.
Show as a table
| Option | minutes | Shape |
|---|---|---|
| Fast | 28 | 5 of 8 done fully, the rest in part |
| Smart | 116 | 8 of 8 done fully |
Pick by task
A simple rule: if you could write the steps yourself, give them to Fast. If you want the agent to find the steps, ask Smart.
| Task | Model |
|---|---|
| Rename a function across a project | Fast |
| Write tests for a module you describe | Fast |
| Summarise a log, a document or a thread | Fast |
| Fix a build that fails with a clear error | Fast |
| Add a feature that touches many files | Smart |
| Find a bug that shows up only sometimes | Smart |
| Review a pull request before you merge | Smart |
| Move a project to a new framework | Smart writes the plan, Fast does the steps |
How the choice spends your hours
Uno AI counts the minutes the AI is working, not the tokens. Both models spend hours at the same rate, so the one that finishes sooner spends less. In our test Fast went through the tasks in about a quarter of the time Smart took, but finished fewer of them fully.
On Pro+AI the difference grows after the 45 full-speed hours: only Fast keeps going. On Max+AI both run with no end. On Small+AI and Plus+AI both stop when the hours end. The honest limits, plan by plan.
Switch per task
In a tool that lets you pick a model, choose it for each chat. Through the API, set the model to uno/fast or uno/smart. A good habit: plan with Smart, then hand the steps to Fast.
# the plan: Smart $ curl https://console.uno.place/v1/chat/completions \ -H "Authorization: Bearer $UNO_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "uno/smart", "messages": [{"role": "user", "content": "Plan the move"}]}' # the steps: Fast $ curl https://console.uno.place/v1/chat/completions \ -H "Authorization: Bearer $UNO_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "uno/fast", "messages": [{"role": "user", "content": "Do step 1"}]}'
Where Fast and Smart fall short
- Both score below the strongest Claude models on the same index: Claude Opus 5.5 has 58, Claude Sonnet 5.5 has 56. For the hardest work, use a premium model or your own Claude or ChatGPT plan.
- Fast finished 5 of 8 tasks fully in our test. The rest needed another pass.
- Smart starts slower: its first text comes after about 4.1 seconds.
- The models behind Fast and Smart can change. The names above are today's.
Both models behind one key, paid in time
Uno AI writes code, builds sites and runs tasks right on your computer. Only the minutes it works count; ten chats in the same minute count as one.
Today: GLM-5.3 Flash (Fast) and MiMo v2.6 Pro (Smart). Dictation into text at no extra cost, within a daily limit.
- 40 hSmall + AIa month
- 100 hPlus + AIa month
- 45 h + FastPro + AIfull speed, then Fast never stops
- UnlimitedMax + AI100 h at full speed, then standard
Pick a computer, then add Uno AI
Every plan is a real Linux computer with root. Add Uno AI to any of them and switch between Fast and Smart per task.
Small
A Telegram bot, a VPN, a small backend
$5/ month
- 1 vCPU · 1 GB
- 15 GB disk
- 25 GB cloud storage
- 5 boosts a month
Plus
Agents that work with your laptop closed
$20/ month
- 4 vCPU · 4 GB
- 40 GB disk
- 100 GB cloud storage
- Uno Work in the cloud
Pro
A product in production, a second MacBook
$70/ month
- 16 vCPU · 16 GB
- 120 GB disk
- 300 GB cloud storage
- Speed of a MacBook Air M4
Max
A team and many agents at once
$200/ month
- 32 vCPU · 48 GB
- 300 GB disk
- 1,024 GB cloud storage
- 5 public IPv4
Add Uno AI to any plan: 40 h of AI work a month on Small, 100 h on Plus, unlimited on Max. Today: GLM-5.3 Flash and MiMo v2.6 Pro. Or sign in with your own Claude or ChatGPT.