Thinkingmachines / inkling / max

T
Thinkingmachines / inkling

max

none low medium high xhigh max
97.6% 199 / 204 fixtures · 1 run(s)
36,058 input / 129,611 total output / 125,628 reasoning within output tokens $0.55854403
62.8% 128 / 204 fixtures · 1 run(s)
23,143 input / 102,793 total output / 99,763 reasoning within output tokens $0.43945465
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-34.8% Text: 97.6% → JSON: 62.8%
+0
Gained
JSON pass / text fail
−71
Lost
Text pass / JSON fail
128
Unchanged Pass
Both pass
5
Unchanged Fail
Both fail
Fixture Reliability Delta
Fixture Text JSON Delta
Benchmark Deltas
Benchmark Text JSON Delta
tag_management 100% 16.7% -83.3%
branch_cleanup 100% 41.7% -58.3%
commit_messages 100% 50% -50%
git_clean 100% 50% -50%
git_log_format 100% 50% -50%
worktree_usage 100% 50% -50%
merge_conflicts 91.7% 50% -41.7%
reflog 100% 58.3% -41.7%
stash_recovery 100% 58.3% -41.7%
cherry_pick 83.3% 50% -33.3%
git_bisect 100% 66.7% -33.3%
rebase 83.3% 50% -33.3%
blame_forensics 100% 83.3% -16.7%
commit_squash 100% 91.7% -8.3%
git_grep 100% 100% + 0%
git_show 100% 100% + 0%
submodule_usage 100% 100% + 0%
Changed Fixtures (71)
f002 -loss f009 -loss f001 -loss f003 -loss f006 -loss f008 -loss f009 -loss f010 -loss f011 -loss f001 -loss f002 -loss f010 -loss f011 -loss f002 -loss f003 -loss f004 -loss f005 -loss f008 -loss f010 -loss f006 -loss f001 -loss f003 -loss f007 -loss f010 -loss f002 -loss f004 -loss f006 -loss f007 -loss f009 -loss f012 -loss f004 -loss f006 -loss f008 -loss f009 -loss f010 -loss f012 -loss f001 -loss f004 -loss f008 -loss f009 -loss f011 -loss f002 -loss f003 -loss f005 -loss f006 -loss f004 -loss f005 -loss f006 -loss f009 -loss f012 -loss f004 -loss f005 -loss f006 -loss f008 -loss f009 -loss f001 -loss f002 -loss f003 -loss f004 -loss f005 -loss f007 -loss f008 -loss f009 -loss f010 -loss f012 -loss f003 -loss f004 -loss f006 -loss f007 -loss f008 -loss f010 -loss
Fixture Gallery (204)