AI Model Reviewer

MiniMax M3

MiniMax (via Fireworks). Run in Raw API call.

64 runs on 59 prompts · 75% finished

Results

The best run of each of its 59 tests first, then every other run. Filter them in the explorer

48 finished runs

  • MiniMax M3, Raw API (one-shot): 004 Brick Breaker NeonLive demo

    004 Brick Breaker Neon

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): 002 Pendulum WaveLive demo

    002 Pendulum Wave

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Codrops-style scroll story: the life of a raindropLive demo

    Codrops-style scroll story: the life of a raindrop

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Artwork style passLive scene

    Artwork style pass

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js)Live scene

    Four-stroke engine cutaway: crank, piston and cam-driven valves (Three.js)

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Hot air balloon festival over rolling fields (Three.js)Live scene

    Hot air balloon festival over rolling fields (Three.js)

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Mechanical beeLive scene

    Mechanical bee

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Synthwave hover-bike racer on a neon grid highway (Three.js)Live scene

    Synthwave hover-bike racer on a neon grid highway (Three.js)

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Outer-space pinball table with planet bumpers (Three.js)Live scene

    Outer-space pinball table with planet bumpers (Three.js)

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • minimax-m3, Raw API (one-shot): Floating pirate island diorama with a cliff cross-section (Three.js)Live scene

    Floating pirate island diorama with a cliff cross-section (Three.js)

    minimax-m3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Sales Dashboard Card

    Sales Dashboard Card

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Cyberpunk Alley

    Cyberpunk Alley

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Corgi Catching a Frisbee

    Corgi Catching a Frisbee

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Chalkboard Coffee Lettering

    Chalkboard Coffee Lettering

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Top-Down City Park

    Top-Down City Park

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Circuit Board Close-Up

    Circuit Board Close-Up

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Chef Flipping a Pan

    Chef Flipping a Pan

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Container Ship

    Container Ship

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Canyon River

    Canyon River

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Calendar Month View

    Calendar Month View

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Butterfly Life Cycle

    Butterfly Life Cycle

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Exploded Burger

    Exploded Burger

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Bubble Tea

    Bubble Tea

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

  • MiniMax M3, Raw API (one-shot): Boxing Ring

    Boxing Ring

    MiniMax M3

    Raw API (one-shot)

    raw API baseline

Numbers: finish rate, cost, wall time, draft scores
Runs
64
on 59 prompts
Finished
75%
48 of 64
Median cost, finished run
not recorded
n=0
Median wall time, finished run
2m 39s
n=48
Recorded spend
$0
$0 on unfinished runs; 64 without a recorded cost

From the numbers

Set against the other models with at least 3 runs (top or bottom third), and its draft-score ranks within single prompts. Nothing here is averaged across prompts.

Strengths

  • Nothing stands out with enough runs to say so.

Weaknesses

  • 48 of 64 runs finished: 18th of 24 models with at least 3 (finish rate)
  • 16 runs ended as the model's own failure (DNF, class MODEL)

Best and hardest run

Best run

Picked as its fastest finished run (no scored result yet).

  • MiniMax M3's output for Breakfast Skillet

    MiniMax M3, raw API call, Breakfast Skillet

    Raw API call, default effort

    raw API baseline

    30 s

    not recorded: the run's token counts were never reported, so there is no estimate

    OK

    Unscored: no rubric for this prompt

Hardest run

Picked as its slowest finished run.

Draft scores by prompt

No scored result yet (only prompts with a published rubric are scored).

By harness

Raw API call644875%not recorded: no finished run with a recorded cost2m 39s n=48
All 59 tests and experiment collections