DeepPlanningdeep-planning · agents
18.0%
ReportedOct 2026
Benchmark results
See the tests behind each score. They use the benchmark’s original scoring scale.
Chat#823.41 ratingEconomics#119.91 ratingInstruction Following#1418.38 ratingSpatial Reasoning#2915.12 ratingStructured Output#1322.06 rating
Agents3 results
t2-bencht2-bench · agents
79.1%
ReportedOct 2026
VITA-Benchvita-bench · agents
29.8%
ReportedOct 2026
Chat2 results
IFEvalifeval · chat
91.5%
ReportedOct 2026
Multi-Challengemultichallenge · chat
54.5%
ReportedOct 2026
General intelligence4 results
C-Evalc-eval · general
88.2%
ReportedOct 2026
Includeinclude · general
75.6%
ReportedOct 2026
MAXIFEmaxife · general
83.4%
ReportedOct 2026
NOVA-63nova-63 · general
55.9%
ReportedOct 2026
Instruction Following1 result
IFBenchifbench · instruction_following
64.5%
ReportedOct 2026
Language5 results
MMLU-Prommlu-pro · language
82.5%
ReportedOct 2026
MMLU-ProXmmlu-prox · language
76.3%
ReportedOct 2026
MMLU-Reduxmmlu-redux · language
91.1%
ReportedOct 2026
MMMLUmmmlu · language
81.2%
ReportedOct 2026
WMT24++wmt24++ · language
72.6%
ReportedOct 2026
Long Context1 result
LongBench v2longbench-v2 · long_context
55.2%
ReportedOct 2026
Math3 results
HMMT 2025hmmt-2025 · math
83.2%
ReportedOct 2026
HMMT25hmmt25 · math
82.9%
ReportedOct 2026
PolyMATHpolymath · math
57.3%
ReportedOct 2026
Reasoning5 results
AA-LCRaa-lcr · reasoning
63.0%
ReportedOct 2026
Global PIQAglobal-piqa · reasoning
83.2%
ReportedOct 2026
GPQAgpqa · reasoning
81.7%
ReportedOct 2026
LiveCodeBench v6livecodebench-v6 · reasoning
65.6%
ReportedOct 2026
SuperGPQAsupergpqa · reasoning
58.2%
ReportedOct 2026
Tool Calling1 result
BFCL-V4bfcl-v4 · tool_calling
66.1%
ReportedOct 2026