Meta's Muse Spark 1.1 scores 48.5 on RadLE 2.0 radiology benchmark

Muse Spark 1.1 is SOTA on the Radiologists Last Exam Handover Readiness Index (RadLE-H), nearing human expert performance. https://twitter.com/drdatta_aiims/status/2076689436092326276

likes: 504
replies: 48
bookmarks: 45
reposts: 40

JR

Jack Rae@jack_w_rae 7 weeks ago

Muse Spark 1.1 is really strong on health topics! I have found this to be the case from personal usage but it’s cool to see it show up on benchmarks also. https://twitter.com/alexandr_wang/status/2076696459005837410

likes: 75
replies: 8
bookmarks: 6
reposts: 2

ZM

Zvi Mowshowitz@TheZvi 7 weeks ago

Love it but also laughing at the idea of a Last Exam 2.0. https://twitter.com/DrDatta_AIIMS/status/2076689408040849507

likes: 408
replies: 20
bookmarks: 27
reposts: 15

RN
Ramez Naam@ramez 7 weeks ago
@TheZvi Last exam v 132.7
- likes: 2
- replies: 0
- bookmarks: 0
- reposts: 0
RG
Riley Goodside@goodside 7 weeks ago
@TheZvi Reverse nominative determinism; same reason there’s lots of Final Fantasy sequels.
- likes: 11
- replies: 0
- bookmarks: 0
- reposts: 0

JW

Jason Wei@_jasonwei 7 weeks ago

Muse Spark 1.1 outperforms GPT-5.6 Sol and Gemini 3.1 on Radiology's Last Exam. We don't beat Fable (yet). And Humans are still a lot better, but we are working on closing the gap! https://twitter.com/DrDatta_AIIMS/status/2076689408040849507

likes: 267
replies: 27
bookmarks: 32
reposts: 32

Meta's Muse Spark 1.1 scores 48.5 on RadLE 2.0 radiology benchmark

Combined views

Meta's Muse Spark 1.1 scores 48.5 on RadLE 2.0 radiology benchmark

Reactions

Combined views