Interaction BenchInteraction Bench
v0.1 preview
LeaderboardEnvironmentsTasksPipelineMethodologyDocsBlogAboutContributors
Github
Kimai

Kimai

Project management
Time tracking for freelancers and teams
Githubkimai/kimai5.1k
Tech stack
PHPSymfonyTwigMySQL
Languages
PHP86.9%
Twig8.7%
JavaScript3.1%
Other1.3%
Available tools
Browser
Terminal
Network
Keep project selections when adding a customer
Keep project selections when adding a customer

Tasks

All 1 task
Keep project selections when adding a customer
#6075 · Base 589296c
Add another customer to my timesheet filter.
50%

Model results

Full leaderboard
Model
Harness
DeepSeek
DeepSeek Flash
DSH
48.4M in · 126.3k out · 48.6M total
DSH · 211 steps · $0.28 · 19m 32s · 48.6M tokens
DSH
211
$0.28
19m 32s
48.4M
126.3k
48.6M
Z.ai
GLM-5.3 Flash
ZCode
27.3M in · 94.9k out · 27.4M total
ZCode · 285 steps · $0.91 · 41m 18s · 27.4M tokens
ZCode
285
$0.91
41m 18s
27.3M
94.9k
27.4M
Claude
Claude Opus 5.5
Claude Code
5.0M in · 33.1k out · 5.0M total
Claude Code · 73 steps · $2.46 · 18m 14s · 5.0M tokens
Claude Code
Claude Code
73
$2.46
18m 14s
5.0M
33.1k
5.0M
MoonshotAI
Kimi K3
Kimi Code
3.5M in · 34.6k out · 3.6M total
Kimi Code · 18 steps · $1.93 · 25m 25s · 3.6M tokens
Kimi
Kimi Code
18
$1.93
25m 25s
3.5M
34.6k
3.6M
OpenAI
GPT-6 Astra
Codex
3.2M in · 7.6k out · 3.2M total
Codex · 54 steps · $4.23 · 7m 56s · 3.2M tokens
Codex
Codex
54
$4.23
7m 56s
3.2M
7.6k
3.2M
Qwen
Qwen3.8 Max
Codex
20.4M in · 83.2k out · 20.4M total
Codex · 193 steps · $6.08 · 1h 03m · 20.4M tokens
Codex
Codex
193
$6.08
1h 03m
20.4M
83.2k
20.4M
Interaction BenchInteraction Bench

Benchmarked on what users feel.

// results

LeaderboardEnvironmentsTasks

// how it works

PipelineMethodologyDocsBlog

// project

AboutContributorsContact usGitHubReport an issue
$ ibench --release 0.1 --statuspreview · built by EvalSys