[ Model ]

Origin. MMLU 81.4%

Our fastest, most responsive model — built for pure speed.

Fast
Efficient
Affordable
Try in QuarkAPI docsPricing
Context
128K

Window

Latency
~18ms

Typical TTFT

Highlight
81.4%

MMLU

API id
origin

Model identifier

Overview

Origin is the foundation of the IdonAI model family — and the default speed path inside Quark. Engineered for low latency, it delivers Flash-class quality on short-form NLP, code assistance, and real-time apps without premium pricing.

Specifications

Context

128K tokens

Latency

~18ms

Pricing

From $0.20 / 1M input tokens

Capabilities

  • 210 tokens/sec throughput
  • JSON and XML structured outputs
  • Strong short-form NLP & coding
  • Flash-tier latency for chat and tools
  • Ethical AI safeguards

Benchmarks

MMLU81.4%

NextGen 90.6%

HumanEval78.9%

NextGen 89.4%

GSM8K91.8%

NextGen 96.2%

MATH74.2%

NextGen 94.1%

Latency

~18ms

Context

128K

Ghost bars show the sibling model. Full Origin ↔ NextGen compare lives on the models index.

Ideal use cases

  • Customer support automation
  • Real-time chat applications
  • Financial data analysis
  • AI-assisted coding
  • Research summarization

Call it

Use model id origin in the chat completions API. Keys are under development — prototype in Quark today.

Path to production

01

Try in Quark

Prompt Origin in the assistant before you wire an integration.

Open Quark →
02

Call the API

Use the same model ID in the unified chat completions API.

Read docs →
03

Scale with pricing

Start free on Quark, then pay-as-you-go or enterprise.

View pricing →

Related models

NextGen

Our most advanced model — designed for the future.

[ Next ]

Run Origin in Quark.

Same model family behind the assistant and the API. Compare on the models index, then ship.

Try QuarkCompare models