Holo3.1: Fast & Local Computer Use Agents

Summary

Holo3.1 is a family of computer-use models released to improve robustness across web, desktop, and mobile environments, agent frameworks, and deployment targets. Based on Qwen, it adds native function-calling alongside structured JSON outputs, expands mobile automation, and offers 0.8B, 4B, 9B, and 35B-A3B sizes for different cost, latency, and performance needs. On AndroidWorld, the 35B-A3B model rises from 67% to 79.3%, while the 4B and 9B variants rise from 58% to 72%; function-calling and native execution achieve near-parity across the stated benchmarks, and Holo3.1 improves more than 25% over Holo3 in the Holotab harness. The release also provides 35B-A3B FP8, Q4 GGUF, and NVFP4 checkpoints for local inference: NVFP4 matches FP8 on OSWorld at about two points below BF16, reaches 1.74 times BF16 token throughput on DGX Spark, and with harness optimizations cuts average step time from 6.8 seconds to 3.3 seconds.

Takeaways

1. 1

   On AndroidWorld, Holo3.1's 35B-A3B model improves from 67% to 79.3%, while the 4B and 9B variants improve from 58% to 72%.

2. 2

   Function-calling and native execution achieve near-parity across OSWorld and the internal benchmark suite, while Holo3.1 improves more than 25% over Holo3 in the Holotab harness.

3. 3

   NVFP4 W4A16 on DGX Spark delivers 1.41× the total token throughput of FP8 and 1.74× that of BF16; combined with harness optimizations, it reduces average step time from 6.8 seconds to 3.3 seconds.