Keep Anthropic. Add every other lab.

Your Anthropic models still have names.

Claude OpusClaude Sonnet

What joining NinjaChat actually adds.

GPT Image 2FLUX 2Imagen 4Nano BananaRecraft V3

Veo 3.1

Seedance 2

GPT-5.4Gemini 3.1Grok 4Kling 2.6

Then these.

1. 01

   Create a key and add credits

   Get an nj_sk_ key at /developers and add a credit pack. Balances are prepaid, and failed generations are refunded.

   verify your key workscopy

   # Sanity check: list every model your key can call (no auth needed)
   curl https://www.ninjachat.ai/api/v1/models
   
   # First authenticated request
   curl https://www.ninjachat.ai/api/v1/chat/completions \
     -H "Authorization: Bearer nj_sk_YOUR_KEY" \
     -H "Content-Type: application/json" \
     -d '{"model": "gpt-5", "messages": [{"role": "user", "content": "Say hi"}]}'

2. 02

   Translate tool use: tool_use / tool_result → tool_calls / role:"tool"

   Anthropic interleaves tool_use and tool_result content blocks inside messages. Chat Completions puts calls on the assistant message (tool_calls, with JSON-string arguments) and results in a dedicated role:"tool" message bound by tool_call_id. Tool schemas also move from input_schema to function.parameters.

   before· anthropiccopy

   tools = [{
       "name": "get_weather",
       "description": "Get weather for a city",
       "input_schema": {
           "type": "object",
           "properties": {"city": {"type": "string"}},
           "required": ["city"],
       },
   }]
   
   msg = client.messages.create(
       model="claude-opus-4-6", max_tokens=1024,
       tools=tools,
       messages=[{"role": "user", "content": "Weather in Tokyo?"}],
   )
   tool_use = next(b for b in msg.content if b.type == "tool_use")
   
   # send the result back as a tool_result content block
   followup = client.messages.create(
       model="claude-opus-4-6", max_tokens=1024, tools=tools,
       messages=[
           {"role": "user", "content": "Weather in Tokyo?"},
           {"role": "assistant", "content": msg.content},
           {"role": "user", "content": [{
               "type": "tool_result",
               "tool_use_id": tool_use.id,
               "content": "22°C, clear",
           }]},
       ],
   )

   →becomes

   after· ninjachatcopy

   tools = [{
       "type": "function",
       "function": {
           "name": "get_weather",
           "description": "Get weather for a city",
           "parameters": {                       # was: input_schema
               "type": "object",
               "properties": {"city": {"type": "string"}},
               "required": ["city"],
           },
       },
   }]
   
   resp = client.chat.completions.create(
       model="claude-opus-4.6",
       tools=tools,
       messages=[{"role": "user", "content": "Weather in Tokyo?"}],
   )
   message = resp["choices"][0]["message"]
   call = message["tool_calls"][0]  # arguments is a JSON string
   
   followup = client.chat.completions.create(
       model="claude-opus-4.6", tools=tools,
       messages=[
           {"role": "user", "content": "Weather in Tokyo?"},
           message,                              # assistant msg w/ tool_calls
           {"role": "tool",                      # was: tool_result block
            "tool_call_id": call["id"],
            "content": "22°C, clear"},
       ],
   )

3. 03

   Streaming: event stream → chat.completion.chunk deltas

   Anthropic streams typed events (content_block_delta, message_delta). Chat Completions streams uniform chunks whose choices[0].delta.content carries the text — most code gets simpler.

   streaming claude with the ninjachat sdkPythonNode.jscopy

   stream = client.chat.completions.create(
       model="claude-sonnet-4.6",
       messages=[{"role": "user", "content": "Explain WebSockets in 3 lines"}],
       stream=True,
   )
   for chunk in stream:
       delta = chunk.get("choices", [{}])[0].get("delta", {}).get("content")
       if delta:
           print(delta, end="", flush=True)

Request translation cheat sheet

Anthropic Messages APINinjaChat Chat Completions

system: "..." (top-level param){role: "system", content: "..."} as the first message

content: [{type: "text", text}] blockscontent as a plain string (arrays accepted for text + image_url parts)

max_tokens (required)max_completion_tokens optional — defaults to 2048, max 131,072

tools[].input_schematools[].function.parameters

tool_use content block (assistant)assistant message .tool_calls[] (arguments = JSON string)

tool_result content block (user){role: "tool", tool_call_id, content} message

stop_reason: "end_turn" / "tool_use" / "max_tokens"finish_reason: "stop" / "tool_calls" / "length"

x-api-key + anthropic-version headersAuthorization: Bearer nj_sk_... (no version header)

Model mapping — Anthropic ids to NinjaChat slugs

Anthropic modelNinjaChat slugPrice / request

claude-opus-4-6claude-opus-4.6$0.050

claude-sonnet-4-6claude-sonnet-4.6$0.030

claude-sonnet-4-5claude-sonnet-4.5$0.030

claude-haiku-4-5claude-haiku-4.5$0.010

What doesn't come along.

FAQ