peak-v4.3
Deep reasoning for hard problems — code, analysis, long-form writing.
- Context
- 65,536 tokens
- Input
- $1.25/MTok
- Output
- $10/MTok
A focused, fast AI assistant. Streaming answers, conversation history, documents, vision, web search — and an OpenAI-compatible API for builders.
Not a demo. The kind of assistant you keep open all day.
Conversations persist across sessions. Pick up where you left off, edit prompts, branch threads.
Ask for a proposal or report. Peak builds it inline, you download it as PDF, Word, or Markdown.
Drop in screenshots, PDFs, spreadsheets, or code. Peak reads them and answers questions about them.
When the answer needs fresh information, Peak pulls live sources and cites them.
Pick the right tool for the job — the same API, the same chat UI.
peak-v4.3
Deep reasoning for hard problems — code, analysis, long-form writing.
peak-v4.3-light
Fast and cheap. The right default for chat, summaries, drafts.
peak-v4.3-vision
Sees images. Use it for screenshots, charts, scanned documents.
Point any OpenAI SDK at https://ai.pearlfibers.com/v1, swap your key, ship. Streaming, usage headers, error shape — all match what your code already expects.
from openai import OpenAI
client = OpenAI(
api_key="sk-peak-…",
base_url="https://ai.pearlfibers.com/v1",
)
print(client.chat.completions.create(
model="peak-v4.3",
messages=[{"role": "user", "content": "Hello!"}],
).choices[0].message.content)
An Iraq-based product team. We use Peak AI ourselves every day.
We don't train on your prompts. See how we handle data →
Sign in once. Your workspace, your billing, one identity.