Skip to content
llmwise
Sign inStart free

Talk to AI

Talk to AI, in a conversation that keeps its thread

Every model in llmwise gets the whole conversation, and you can switch between 15 of them mid-chat without starting over. It's a text chat, with no voice mode: type, and read the answer.

A real answer from our test runs

GPT-6 Luna

In two paragraphs of at most 90 words each, first argue that a city should make its buses free, then make the strongest case against it. The second paragraph must be as persuasive as the first. No headings.

GPT-6 Luna · passed our check · September 27, 2026

Making buses free would give every resident reliable access to jobs, school, health care, and daily necessities, regardless of income. It would also speed boarding, reduce fare-enforcement costs, and encourage people to leave cars at home—easing congestion, pollution, and parking pressure. A city that treats public transit as essential infrastructure, rather than a service people must qualify to afford, can build a more connected and equitable community.

But “free” buses still have to be paid for, and the bill could divert money from more urgent improvements: frequent service, safer stops, cleaner vehicles, and routes to underserved neighborhoods. If buses remain slow, crowded, or unreliable, removing fares may do little to persuade drivers to switch, while reducing funds that could make transit genuinely useful. Targeted fare assistance can protect riders who need it most without subsidizing every trip, including those by people who can readily afford to pay.

Ask your own question: sign up free for 5 messages on 12 of the models, no card.

Start free

Test runs and prices checked . Updated .

Short answer

Yes, in writing: every message goes to the model with the whole conversation, up to 200k tokens, and you can switch models without losing it. There's no voice mode.

How a conversation works

What happens to a conversation as it grows.

  • The whole thread, every time

    Each reply is written with everything said so far, so you can say “shorter” or “go back to the second idea” and be understood.

  • Change who you're talking to

    Pick another model for the next message: it reads the same thread, so a second opinion needs no copy and paste.

  • Long conversations

    Past 64k tokens a reply counts as 2 messages, past 128k as 4, and at 200k the chat is full; start a new one then.

  • Memory across chats

    Save what the AI should know about you: 20 memories on Free, 100 on a paid plan, for every chat.

How much each model can read

How much of a conversation, with its files, each model can read at once.

  • 1.05M tokens

    GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, Gemini 3.1 Pro (preview), Gemini 3.8 Flash, DeepSeek V4.1 Flash, DeepSeek V4 Pro, Kimi K3, GLM 5.3, and GLM 5.3 Flash

  • 1M tokens

    Claude Fable 5.1, Claude Opus 5.5, and Claude Sonnet 5

  • 500K tokens

    Grok 4.7

  • 200K tokens

    Claude Haiku 4.5

Models that get the tone right

Our writing and customer-support prompts are graded for tone and clarity as well as content, by a fixed rubric; here's how the models did on those 10.

Models ranked on writing and support prompts
#ModelPassedHard onesFree trialOn ProCost per reply
1GPT-6 LunaOpenAI9 of 104 of 4Yes60 a day$0.0001
2DeepSeek V4.1 FlashDeepSeek9 of 104 of 4Yes60 a day$0.0003
3Claude Sonnet 5Anthropic9 of 104 of 4Yes125 a month$0.0035
4GLM 5.3Z.ai9 of 103 of 4Yes250 a month$0.0005
5Gemini 3.1 Pro (preview)Google9 of 103 of 4Yes125 a month$0.0099
6Claude Opus 5.5Anthropic9 of 103 of 4No62 a month$0.0146
7GPT-6 AstraOpenAI8 of 104 of 4No31 a month$0.0113
8Kimi K3Moonshot8 of 102 of 4Yes125 a month$0.0042
9Grok 4.7xAI7 of 104 of 4Yes250 a month$0.0045
10GPT-6 SolOpenAI7 of 104 of 4Yes125 a month$0.0026
11Gemini 3.8 FlashGoogle7 of 103 of 4Yes250 a month$0.0015
12DeepSeek V4 ProDeepSeek7 of 102 of 4Yes250 a month$0.0008
13Claude Haiku 4.5Anthropic7 of 102 of 4Yes250 a month$0.0013
14Claude Fable 5.1Anthropic7 of 101 of 4No31 a month$0.0197
15GLM 5.3 FlashZ.ai4 of 101 of 4Yes60 a day$0.0003
10 prompts per model (writing and customer support), run on September 27, 2026. Ranked by how many prompts each model passed, then how many of the hard ones, then by the smaller message (the everyday models first), then by the lower cost per reply. No ranking is chosen by hand. Cost per reply is what OpenRouter charged us on average; in llmwise you pay per message.

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.

More ways to chat

The other chat pages, each with its own models and test results.

Questions

Can I talk to the AI by voice?

No: llmwise is a text chat, with no voice mode. You type, and the model answers in writing.

Does the AI remember what I said earlier?

Within a chat, yes: every message goes to the model with the whole conversation. Across chats, memory keeps the facts you save (20 on Free, 100 on a paid plan).

How long can a conversation be?

Up to 200k tokens. Past 64k tokens each reply counts as 2 messages, past 128k as 4, and the message box says so before you send.

Can I delete what I said?

You can delete a chat, or every chat, at any time. The Privacy Policy lists where each model's messages go.

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.

Start freeSee pricing