Hanzo / Enso
The right intelligence.Every time.
One endpoint for every frontier model. Enso routes each request for quality, speed and cost.
The default for everyone, Free plan included: it picks the tier for each request and bills at the tier that answers.
OpenAI-compatible APIAnthropic-compatible API
Enso Flash
enso-flash1M context$0.33 in · per 1M tokens $1.65 out
For chat, extraction, classification and simple agent steps at volume.
Enso Pro
enso-pro262K context$4.40 in · per 1M tokens $22 out
For coding, code review and agents that answer while someone waits.
Enso Ultra
enso-ultra1M context$5.50 in · per 1M tokens $27.50 out
For work where a wrong answer is expensive: research, security review and long autonomous runs. For hard reasoning over long material: it thinks at high effort, holds a 1M-token context and reads images.
One endpoint for every frontier model.
Enso routes each request for quality, speed and cost.
Requests
- researchEnso Ultra
- codingEnso Pro
- chatEnso Flash
- Enso Flashcost · at volume
enso-flash$0.33 in · per 1M tokens $1.65 out 1M context
- Enso Prospeed · while someone waits
enso-pro$4.40 in · per 1M tokens $22 out 262K context
- Enso Ultraquality · a wrong answer is expensive
enso-ultra$5.50 in · per 1M tokens $27.50 out 1M context
Send a request
Frontier AI for everything you build.
Language, routing, decisions and creative models.
- Flagship modelEnsoOrchestrates every model for higher intelligence, faster and at lower cost.
- KaiPrecise decisions and structured reasoning.
- ZenDeep reasoning for complex problems.
- IroImage generation and visual understanding.
- KomaVideo generation and world simulation.
- OtoSound generation and audio understanding.
- KoeVoice synthesis and speech understanding.
- UtaMusic generation and composition.
- Mori3D generation and spatial understanding.
A New Pareto Frontier of Intelligence
GPQA Diamond challenges AI with 198 graduate-level science questions. See how Enso balances accuracy and cost across 1,000 answers.
| Model | GPQA Diamond % | $ per 1,000 questions |
|---|---|---|
| Enso Ultra | 98.0% | $14.76 |
| Enso | 96.5% | $10.03 |
| Enso Flash | 95.5% | $3.26 |
| GPT-6 Sol ↗ | 94.3% | $5.09 |
| Kimi K3 ↗ | 93.5% | $10.61 |
| Grok 4.7 ↗ | 92.7% | $4.57 |
| GPT-5.6 Sol | 91.9% | $10.08 |
| GLM 5.3 ↗ | 90.9% | $0.945 |
| Opus 5.5 | 90.9% | $21.23 |
| Sonnet 5.5 | 90.4% | $10.41 |
| GPT-5.5 | 90.4% | $18.64 |
| Fable 5.1 | 87.9% | $46.25 |
| Opus 4.8 | 87.4% | $45.01 |
| Haiku 5.5 | 83.3% | $1.07 |
| Fable 5 | 82.8% | $14.95 |
| Sonnet 5 | 77.8% | $42.91 |
| DeepSeek V4 Pro | 76.3% | $1.05 |








