Solving Hack The Box Challenges with Grok 4.6 | TheArtificialQ Blog
When I tested an older version of Grok this spring, my only comment was blunt: “grok-4.20 was useless.” So when Grok 4.5 was released a month ago and started receiving very positive reviews, I was quite surprised.
Today, Grok 4.6 was released and I’ve finally had a chance to test it against HTB challenges, and the team behind it deserves an apology. I’m not sure what the folks at xAI did, but the performance increase over the older version is jaw-dropping.
This blog post is part of a series of tests for the HTB-Challenger Benchmark.<br>See the benchmark results page for all results and the benchmark methodology to learn how the benchmark is calculated.
During my testing, I watched this model crack one HTB challenge after another in a steady, confident way, and the final boss fight against the last two challenges was exhilarating, even though it was ultimately unsuccessful. In the end, with a 80% score, it’s the undisputed new king of my leaderboard - not only because of its results, but also because of its speed (it had the fewest steps per challenge of all the models I’ve tested so far) and cost. Look at the Cost vs. Benchmark Score graph on the HTB-Challenger Benchmark page to see what I mean.
So let’s forget that useless old version. For me, Grok 4.6 is currently the best model I’ve tested on offensive-security challenges.
Overall benchmark results
Number of challenges: 16
Number of solved challenges: 14
Number of false positives: 0
Runs where the model gave up: 0
Runs that reached the step or cost limit: 2
Runs where the model got stuck: 0
Benchmark score: 80.0%
Metric<br>Per challenge (median)<br>Total
Model steps<br>14<br>452
Model cost<br>$0.23<br>$17.31
Duration<br>00:03:38<br>02:45:35
Number of input tokens<br>0.23M<br>20.29M
Number of output tokens<br>0.01M<br>0.47M
Number of read_file tool calls<br>1.0<br>41
Number of write_file tool calls<br>0.0<br>35
Number of execute_command tool calls<br>10.0<br>403
Number of web_search tool calls<br>0.0<br>15
Results by challenge difficulty
Very Easy challenges
Number of challenges: 4
Number of solved challenges: 4
Number of false positives: 0
Runs where the model gave up: 0
Runs that reached the step or cost limit: 0
Runs where the model got stuck: 0
Benchmark score: 100.0%
Metric<br>Per challenge (median)<br>Total
Model steps<br>11.5<br>50
Model cost<br>$0.10<br>$0.96
Duration<br>00:01:26<br>00:09:26
Number of input tokens<br>0.10M<br>1.05M
Number of output tokens<br>0.00M<br>0.03M
Number of read_file tool calls<br>0.5
Number of write_file tool calls<br>0.0
Number of execute_command tool calls<br>9.0<br>42
Number of web_search tool calls<br>0.5
Easy challenges
Number of challenges: 4
Number of solved challenges: 4
Number of false positives: 0
Runs where the model gave up: 0
Runs that reached the step or cost limit: 0
Runs where the model got stuck: 0
Benchmark score: 100.0%
Metric<br>Per challenge (median)<br>Total
Model steps<br>11.5<br>43
Model cost<br>$0.14<br>$0.58
Duration<br>00:03:00<br>00:12:29
Number of input tokens<br>0.12M<br>0.51M
Number of output tokens<br>0.01M<br>0.03M
Number of read_file tool calls<br>1.0
Number of write_file tool calls<br>2.0
Number of execute_command tool calls<br>8.0<br>30
Number of web_search tool calls<br>0.5
Medium challenges
Number of challenges: 4
Number of solved challenges: 4
Number of false positives: 0
Runs where the model gave up: 0
Runs that reached the step or cost limit: 0
Runs where the model got stuck: 0
Benchmark score: 100.0%
Metric<br>Per challenge (median)<br>Total
Model steps<br>19<br>106
Model cost<br>$0.58<br>$3.98
Duration<br>00:09:39<br>00:45:15
Number of input tokens<br>0.61M<br>4.50M
Number of output tokens<br>0.02M<br>0.11M
Number of read_file tool calls<br>0.0
Number of write_file tool calls<br>0.5
Number of execute_command tool calls<br>16.5<br>102
Number of web_search tool calls<br>0.0
Hard challenges
Number of challenges: 4
Number of solved challenges: 2
Number of false positives: 0
Runs where the model gave up: 0
Runs that reached the step or cost limit: 2
Runs where the model got stuck: 0
Benchmark score: 50.0%
Metric<br>Per challenge (median)<br>Total
Model steps<br>63.5<br>253
Model cost<br>$2.89<br>$11.79
Duration<br>00:24:56<br>01:38:25
Number of input tokens<br>3.50M<br>14.23M
Number of output tokens<br>0.08M<br>0.31M
Number of read_file tool calls<br>4.5<br>25
Number of write_file tool calls<br>1.0<br>20
Number of execute_command tool calls<br>55.0<br>229
Number of web_search tool calls<br>1.0