Introducing Holo4: powering generalist computer-use agents

Models API

Multimodal models for real-world use

Process text, images, and documents, with computer-use capabilities for any digital environment.

FastNew

Holo4-35B-A3B

holo4-35b-a3b

Next-generation MoE with a longer context window. Near-flagship accuracy at low cost and latency, built for computer-use.

OSWorld
80.8%Desktop tasks
Context
262KTokens
Input
$0.30$0.03 cached
Output
$2.00Per 1M tokens

MoE, 35B total, 3B active

FlagshipNew

Holo4-27B

holo4-27b

Dense flagship-class reasoning for complex multi-step tasks across web, desktop, and mobile, with an extended context window.

OSWorld
85.2%Desktop tasks
Context
262KTokens
Input
$0.40$0.04 cached
Output
$3.00Per 1M tokens

Dense, 27B

Previous generation, still served

ModelContextInputOutput
Holo3-1-35B-A3B
holo3-1-35b-a3b
65,536$0.25$1.80
Holo3-122B-A10B
holo3-122b-a10b
65,536$0.40$3.00

Rate-limited free access to Holo3-1-35B-A3B. No credit card required. Pay-as-you-go for every model, with higher rate limits.

After you pay, it can take up to 15 minutes for the system to process your payment and credit your balance.

Frequently asked questions

Start with Holo4. Holo4-35B-A3B is the default for production computer-use workloads: longer context, strong accuracy, and low cost and latency. For denser flagship-class reasoning on complex multi-step tasks, use Holo4-27B. Earlier Holo3 models remain available on the same Models API if you need them. Switching only requires a model ID change.

Multimodal models for real-world use

Process text, images, and documents, with computer-use capabilities for any digital environment. Rate-limited free tier, pay-as-you-go beyond.