Unverified50% confidenceFactExact time
In the smart-home scenario tool-calling test, Audio 2B got 9/10 right and Omni 2B got 10/10 right; both correctly handled negative instructions and multi-action commands, and neither wrongly called tools when it could not complete the task.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/4/2026
First Seen
Valid until: 2/2/2027
Sources
Running Audio and Vision Locally with a 2B Model: LFM3 Hands-On Revealed
bilibili枫叶边城11/4/2026
Related Entities
Related Claims
UnverifiedWhen screening AI music, you should listen once each on speakers and headphones, and skip the opening ten-second 'show-off' section to judge overall usability.66% similarUnverifiedThe plugin features distinct notification sounds for four event types: task completion, response needed, authorization needed, and runtime error.66% similarUnverifiedThe Unboom workflow includes a scan mode (playing a test signal through the Mac microphone and capturing the room response to automatically locate problem frequencies) and a manual mode (directly selecting frequencies by ear); after selection, the attenuation strength can be chosen from three levels—Gentle / Medium / Strong—and finally the Correction switch is turned on to take effect.64% similarUnverifiedRelying purely on acoustic models without combining task context (such as a spelling task), speech recognition errors are difficult to avoid.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134837API
curl https://kongchang.com/api/v1/knowledge/claims/1134837MCP
get_claim(id=1134837)