Plotly tests M3.1 Flash on data analytics: ties Opus 5.5 at 1/3 price but 6x slower

Plotly's hands-on benchmark found M3.1 Flash ties Opus 5.5 and Argon on data-analytics tasks at about one-third the price, but burns so many tokens it takes 3 minutes per question versus 30 seconds.

Chris Parmer

@chriddypArchived 8 days agoOriginally posted 2026.10.20

Plotly tests M3.1 Flash on data analytics: ties Opus 5.5 at 1/3 price but 6x slower (image 1 of 3)

Plotly tests M3.1 Flash on data analytics: ties Opus 5.5 at 1/3 price but 6x slower (image 2 of 3)

Plotly tests M3.1 Flash on data analytics: ties Opus 5.5 at 1/3 price but 6x slower (image 3 of 3)

Chris Parmer @chriddypOriginally posted 2026.10.20

Open original on X

Trying out MiniMax M3.1 for data analytics today. Best Chinese model so far, ties Opus 5.5 & Argon at 1/3rd the price. However, it uses a loooot of tokens which makes it the slowest model we've tested so far: 3 min to answer each question vs 30s. https://plotly.com/blog/minimax-m3-1-flash-plotly-data-analytics-bench/

x.com/chriddyp/status/2112412421205991541Archived 2026.10.20 20:26

Open original on X