We use cookies to enhance your experience, provide personalised content, and analyse our traffic; see our Cookie Policy.
anthropic/claude-sonnet-5A Sonnet built for agents, with a 1M-token context
According to Anthropic, Claude Sonnet 5 was the most agentic Sonnet yet when it launched on 30 June 2026: it plans, uses tools and keeps its thread on long tasks. Anthropic now lists it as legacy: Sonnet 5.5 is the current Sonnet.
It is still a very good everyday model: customer assistants, code review, structured extraction, long-document summaries, with a 1M-token context, images and PDFs. According to Anthropic, its performance approaches Opus 4.8 and it hallucinates less than Sonnet 4.6.
On Tchavi, one API key opens 100+ models: compare Sonnet 5, Sonnet 5.5 and Opus by changing the model field. You pay with mobile money, no card and no KYC.
Plans, uses tools and works for long stretches without losing the thread, per Anthropic
The same context window as the Opus models
On by default; it can be switched off for faster replies
Reads images and PDFs in the same request as your text
tool_choice "any" or named is accepted, which Sonnet 5.5 rejects
Room for long files, full reports and large diffs
Turn this rambling meeting transcript into a decision log: what was decided, by whom, by when, and what is still open.
Read this PDF invoice and return the line items as JSON (description, quantity, price, total), flagging anything that does not add up.
You have the tools search_orders and refund_order. A customer writes: "My order never arrived". Chain the calls you need, then reply to them in three sentences.
Sonnet 5.5 for a new project: it is Anthropic's current Sonnet. Keep Sonnet 5 if your code forces a specific tool (tool_choice) or turns thinking off: Sonnet 5.5 rejects both with a 400 error.
Sonnet 5 for everyday assistants and agent loops. Opus 5.5, which Anthropic aims at long-running agentic coding, when a mistake is expensive: repository refactors, security reviews.
Yes: tool_choice "any" or a named tool is accepted on Sonnet 5. Sonnet 5.5 and Opus 5.5 return a 400 error in that case.
1M tokens in and up to 128K out, with images and PDFs.
With the OpenAI SDK, change the base URL, use your sk-tch-… key and model: "claude-sonnet-5" on /v1/chat/completions.
Yes. Top up your Tchavi account with Orange Money, MTN MoMo, M-Pesa or another mobile money operator. No bank card is needed. Credits land as soon as the payment is confirmed and are deducted as you use them.
Yes. Keep your current OpenAI SDK, just change the base URL to Tchavi's (shown in the API tab) and use your Tchavi key. The rest of your code stays the same.
import Tchavi from '@tchavi/sdk';
const client = new Tchavi({ apiKey: 'API_KEY' });
const response = await client.chat.completions.create({
model: 'claude-sonnet-5',
messages: [
{
"role": "user",
"content": "Hello!"
}
],
});
console.log(response.choices[0].message.content);
console.log('Credits used:', response.tchavi.credits_used);Every parameter this model accepts on the API. Check the SDK for full typings.
modelRequiredID of the chat model (e.g. `claude-sonnet-5`).
messagesRequiredConversation messages. Each item has a `role` (system/user/assistant) and a `content` that's either a plain string or an array of content blocks (`text`, `image_url`, `document_url`).
temperature1Sampling temperature between 0 and 2. Higher values = more random.
max_tokensMaximum number of tokens to generate in the response.
streamfalseIf true, partial message deltas are streamed as Server-Sent Events.
toolsFunction-calling toolbox. Pass an array of `{ type: "function", function: { name, description, parameters } }`. Returned tool calls land in `choices[].message.tool_calls`. Only models with `supportsTools: true` accept this.
tool_choiceHow the model picks among tools: `"auto"` (default), `"none"` to force a plain text reply, or `{ type: "function", function: { name } }` to force a specific tool. Only with `supportsTools: true`.
top_p1Nucleus sampling: only tokens in the top `top_p` probability mass are considered.
stopUp to 4 sequences where the API will stop generating further tokens.
frequency_penalty0Between -2.0 and 2.0. Positive values penalize tokens that have appeared frequently.
presence_penalty0Between -2.0 and 2.0. Positive values encourage the model to talk about new topics.
messages[].content[].image_urlAttach an image on a user message. Pass a content block `{ "type": "image_url", "image_url": { "url": ... } }`. PNG / JPEG / WebP, up to 4 per request. Only models with `supportsImageInput: true` accept this block (e.g. `gpt-4o`, `claude-opus-4-7`).
messages[].content[].document_urlAttach a PDF on a user message. Pass a content block `{ "type": "document_url", "document_url": { "url": ... } }`. Up to 5 PDFs per request. Only models with `supportsDocuments: true` accept this block (e.g. `claude-sonnet-4-6`, `gpt-4.1`).
effort"high"Reasoning depth on Anthropic adaptive-thinking models (Opus 4.7+ / Sonnet 4.6+). Values `"low"`, `"medium"`, `"high"`. Higher = more deliberate, slower. Defaults to `"high"` on Opus 4.8.
Estimated over 30 days, for a typical message: ~500 input + ~150 output tokens, billed to the thousandth of a credit.