Kimi Code CLI
1 model ran in it: Kimi K3.
15 runs on 15 prompts · 93% finished
Results
The best run of each of its 15 tests first, then every other run. Filter them in the explorer
14 finished runs
Live scene
Folding road
Kimi K3
Kimi Code CLI 2.1.0
skills off
The run did not publish vendor/three.min.js; it cannot run as built without them, so it is not played here.
Numbers: finish rate, cost, wall time, draft scores
- Runs
- 15
- on 15 prompts
- Finished
- 93%
- 14 of 15
- Median cost, finished run
- $1.30
- n=14
- Median wall time, finished run
- 22m 50s
- n=14
- Recorded spend
- $35.36
- $2.29 on unfinished runs
From the numbers
Set against the other harnesses with at least 3 runs (top or bottom third), and its draft-score ranks within single prompts. Nothing here is averaged across prompts.
Strengths
- 14 of 15 runs finished: 3rd of 14 harnesses with at least 3 (finish rate)
- Draft score 9.5/10 on Animated pelican riding a bicycle (SMIL/CSS) (Kimi K3): ranked 2nd of 12
Weaknesses
- Draft score 8/10 on MacBook Pro 16" realistic 3D render (Playcode benchmark) (Kimi K3): ranked 9th of 12
- Draft score 7.5/10 on Lighthouse on a headland at night (Kimi K3): ranked 9th of 12
- Other unfinished runs, not counted against the model: 1 TIMEOUT
Best and hardest run
Best run
Picked as its best draft-score rank on a prompt.
-
Kimi K3, Kimi Code CLI, Animated pelican riding a bicycle (SMIL/CSS)
Kimi Code CLI 2.1.0, max effort, via Moonshot
skills off
13m 32s
$0.74 estimated
OK
AI judge 9.5/10 Draft scorerubric
Hardest run
Picked as its costliest run that did not finish.
-
Did not finish · re-run queued
Kimi K3, Kimi Code CLI, gpu_fluid_sim
Kimi Code CLI 2.1.0, max effort, via Moonshot
skills off
Stopped by us
57m 02s $2.29 estimated
Draft scores by prompt
| Isometric SVG town with sprite sheet | Kimi K3 | 9/10 draft | 8th of 11 |
|---|---|---|---|
| Lighthouse on a headland at night | Kimi K3 | 7.5/10 draft | 9th of 12 |
| MacBook Pro 16" realistic 3D render (Playcode benchmark) | Kimi K3 | 8/10 draft | 9th of 12 |
| Pelican riding a bicycle (control) | Kimi K3 | 8.5/10 draft | 9th of 13 |
| Animated pelican riding a bicycle (SMIL/CSS) | Kimi K3 | 9.5/10 draft | 2nd of 12 |
By model
| Kimi K3 | 15 | 14 | 93% | $1.30 n=14 | 22m 50s n=14 |
|---|
All 15 tests
- Animated pelican riding a bicycle (SMIL/CSS) (1 run)
- Isometric SVG town with sprite sheet (1 run)
- Pelican riding a bicycle (control) (1 run)
- MacBook Pro 16" realistic 3D render (Playcode benchmark) (1 run)
- Lighthouse on a headland at night (1 run)
- Pelican car france (1 run)
- Dolphin vlog bicycle (1 run)
- Fjord scene (Three.js) (1 run)
- Whale clouds (1 run)
- California brown pelican riding a bicycle (detailed spec) (1 run)
- BMW M4 Competition, side view (flat vector) (1 run)
- Xbox 360 controller (1 run)
- DualSense controller, max fidelity (1 run)
- GPU fluid sim (1 run)
- Folding road (1 run)