Z-ai / glm-5.2 / medium

Z
Z-ai / glm-5.2

medium

none low medium high xhigh
95.1% 194 / 204 fixtures · 1 run(s)
36,570 input / 67,694 total output / 58,657 reasoning within output tokens $0.32963324
74.5% 152 / 204 fixtures · 1 run(s)
38,838 input / 50,035 total output / 44,719 reasoning within output tokens $0.23871336
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-20.6% Text: 95.1% → JSON: 74.5%
+3
Gained
JSON pass / text fail
−45
Lost
Text pass / JSON fail
149
Unchanged Pass
Both pass
7
Unchanged Fail
Both fail
Fixture Reliability Delta
Fixture Text JSON Delta
Benchmark Deltas
Benchmark Text JSON Delta
commit_squash 91.7% 50% -41.7%
merge_conflicts 83.3% 41.7% -41.7%
branch_cleanup 100% 66.7% -33.3%
reflog 100% 66.7% -33.3%
submodule_usage 100% 66.7% -33.3%
git_grep 100% 75% -25%
tag_management 100% 75% -25%
blame_forensics 100% 83.3% -16.7%
git_bisect 100% 83.3% -16.7%
git_show 91.7% 75% -16.7%
rebase 91.7% 75% -16.7%
stash_recovery 100% 83.3% -16.7%
cherry_pick 83.3% 66.7% -16.7%
commit_messages 100% 91.7% -8.3%
git_clean 100% 91.7% -8.3%
git_log_format 100% 100% + 0%
worktree_usage 75% 75% + 0%
Changed Fixtures (48)
f005 -loss f008 -loss f002 -loss f003 -loss f011 -loss f012 -loss f010 -loss f011 -loss f005 -loss f001 -loss f003 -loss f004 -loss f007 -loss f010 -loss f002 -loss f007 -loss f010 -loss f006 -loss f008 -loss f009 -loss f002 +gain f003 -loss f005 -loss f006 -loss f002 -loss f003 -loss f005 -loss f006 -loss f007 -loss f001 -loss f011 -loss f002 -loss f004 -loss f006 -loss f012 -loss f007 -loss f011 -loss f002 -loss f005 -loss f006 -loss f008 -loss f001 -loss f005 -loss f006 -loss f005 -loss f007 +gain f011 -loss f012 +gain
Fixture Gallery (204)