DrivingBench

Equal contribution

Can frontier models drive a real car? We give them control of a Toyota Corolla’s steering, accelerator, and brakes, then evaluate them on a fixed cone course.

Follow @drivingbench Paper Eval traces
0:00 / 1:08

Leaderboard

Up to 3 attempts in one continuous chat. Expand a model to inspect its runs, then click a run to view its trace and video.

GPT-6 Astra
Codex · medium
100%
in5:22
49%100%5:22—
ExpandCollapse
attemptprogressHow far along the course centerline the attempt got while staying within 4 m of it, as a share of the centerline length to the finish zone; a collision keeps the progress reached before it.distanceGPS speed integrated from the first accepted set_motion to the end of the attempt's last engagement.finish timeFirst accepted set_motion to the end of the attempt's last engagement. Shown for completed runs only; DNF means did not finish.commandsAccepted set_motion and stop_now calls during the attempt.tokens · costTotal tokens and cost at list prices for the attempt, including the reflection after it.
same chat
#1
49%
67.3 mDNF81.2M · $2.01
#2
100%
134.7 m5:22246.6M · $7.74
Claude Fable 5.1
Claude Code · medium
45%
9%10%45%
ExpandCollapse
attemptprogressHow far along the course centerline the attempt got while staying within 4 m of it, as a share of the centerline length to the finish zone; a collision keeps the progress reached before it.distanceGPS speed integrated from the first accepted set_motion to the end of the attempt's last engagement.finish timeFirst accepted set_motion to the end of the attempt's last engagement. Shown for completed runs only; DNF means did not finish.commandsAccepted set_motion and stop_now calls during the attempt.tokens · costTotal tokens and cost at list prices for the attempt, including the reflection after it.
same chat
#1
9%
17.5 mDNF3581k · $0.96
#2
10%
27.3 mDNF4866k · $1.35
#3
45%
73.7 mDNF82.1M · $1.64
Grok 4.6
Cursor · medium
11%
8%11%10%
ExpandCollapse
attemptprogressHow far along the course centerline the attempt got while staying within 4 m of it, as a share of the centerline length to the finish zone; a collision keeps the progress reached before it.distanceGPS speed integrated from the first accepted set_motion to the end of the attempt's last engagement.finish timeFirst accepted set_motion to the end of the attempt's last engagement. Shown for completed runs only; DNF means did not finish.commandsAccepted set_motion and stop_now calls during the attempt.tokens · costTotal tokens and cost at list prices for the attempt, including the reflection after it.
same chat
#1
8%
14.4 mDNF2216k · $0.18
#2
11%
22.6 mDNF3294k · $0.19
#3
10%
22.2 mDNF3516k · $0.29
GPT-5.6 Sol
Codex · medium
6%
6%6%6%
ExpandCollapse
attemptprogressHow far along the course centerline the attempt got while staying within 4 m of it, as a share of the centerline length to the finish zone; a collision keeps the progress reached before it.distanceGPS speed integrated from the first accepted set_motion to the end of the attempt's last engagement.finish timeFirst accepted set_motion to the end of the attempt's last engagement. Shown for completed runs only; DNF means did not finish.commandsAccepted set_motion and stop_now calls during the attempt.tokens · costTotal tokens and cost at list prices for the attempt, including the reflection after it.
same chat
#1
6%
15.0 mDNF3364k · $0.35
#2
6%
15.6 mDNF4729k · $0.43
#3
6%
17.1 mDNF2517k · $0.27

Questions? Want your model evaluated next? Reach out at team@drivingbench.com

Explore the trackView the trajectory replays

Explore the course