Skip to content
BetaBenchAlert v0.1 is in Beta.Numbers are real, but pages and rules can still change.See what changed

Grok Build

Grok 4.6

Inside Grok Build we ask for grok-4.6 every hour. Everything below comes from those runs on our lab machine.

Grok Buildgrok-4.6Think setting xhighLast round just nowFeatured

Write speed, last 24 hours

37.0answer tok/s

6.46s typical total wait

Likely middle: 6.32s6.89s

Visible answer tokens per second across the whole command. Hidden thinking does not make this number larger. Higher is faster. These are the middle values of 25 completed checks out of 24 scheduled in the last 24 hours.

Time to first word
3.62s
How long before the first visible word shows up. Lower is better.
Typical total wait
6.46s
How long the whole command took, start to finish, including the app booting. Lower is faster.
Last ping
6.61s
The newest tiny “reply ok” check. It measures wake-up time, not writing speed.

What stands out

Four facts pulled straight from the saved runs.

Board position
#2 of 10
Just ahead: Grok 4.5 at 6.26s total wait. Just behind: Sonnet 5 at 7.00s total wait.
Answered when scheduled
25/24
Completed answers out of all scheduled checks in the last 24 hours.
Served as
grok-4.6-build
The model name the app reported back. If it differs from the pin, we say so.
Round-to-round swing
3.0×
How much the fastest round beat the slowest one. A big swing means a noisy day.

Recent rounds

Newest first. Fails stay in the list with their reason.

WhenResultWriting speedTotal wait
just nowokplugins=2:codex,cloudflare,hooks=1; no tool calls33.0 tok/s7.25s
1h agookplugins=2:codex,cloudflare,hooks=1; no tool calls36.0 tok/s6.63s
1h agookplugins=2:codex,cloudflare,hooks=1; no tool calls39.1 tok/s6.11s
3h agookplugins=2:codex,cloudflare,hooks=1; no tool calls28.0 tok/s8.53s
3h agookplugins=2:codex,cloudflare,hooks=1; no tool calls40.1 tok/s5.96s
5h agookplugins=2:codex,cloudflare,hooks=1; no tool calls29.6 tok/s8.08s
6h agookplugins=2:codex,cloudflare,hooks=1; no tool calls39.2 tok/s6.10s
7h agookplugins=2:codex,cloudflare,hooks=1; no tool calls37.2 tok/s6.43s

All rounds are in Logs.

Read next

Compare with Fable 5

Its closest rival in Claude Code — same window, same units, with the leader marked on every row.

Compare with Grok 4.5

The pin next to it in Grok Build. See what the tier jump buys inside one app.

More models

How to read this

Every number here comes from our lab machine running the paid app, one try per hourly round. The live check uses test kit v0.2; recovered history appears only after its raw file passes the same visible-speed rules. Our measurements, not the vendor’s.