Benchmark Almanac

Inkling

Inkling ranks first on StrongREJECT.

Thinking Machines Released 15 Jul 2026
88
Benchmarks with a score
1
Ranked first, out of five or more models
3
In the top three
Thinking Machines Jul 2026

Inkling

Open weights

148.5 #70 of 271
Epoch Capabilities Index
  • IFBench 79.8% #9 /70
  • MMMU-Pro 73.5% #39 /72
  • τ³-bench Banking 25% #18 /25

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
IFBench 79.8% #9 of 70 Model card
LMArena (Elo) 1,442 #72 of 333 Official leaderboard
Longform Creative Writing (EQ-Bench) 72.5 #35 of 119 Official leaderboard
Humanity's Last Exam 46% #48 of 147 Model card
Global MMLU Lite 3.0 88.7% #1 of 3 Model card
Creative Writing v3 70.9 #49 of 120 Official leaderboard
SEA-HELM 71.3 #30 of 62 Official leaderboard
EQ-Bench 4 1,212 #16 of 32 Official leaderboard
SimpleQA Verified 40.3% #51 of 84 Epoch AI

Math and science

BenchmarkScoreRankSource
AIME 2026 97.1% #5 of 34 Model card
GPQA Diamond 88.3% #54 of 219 Epoch AI
OTIS Mock AIME 2024-2025 88.9% #65 of 202 Epoch AI
DTBench 79.1% #63 of 178 Official leaderboard
CritPt 5.43% #68 of 160 Epoch AI
ARC-AGI-2 36.5% #46 of 95 Official leaderboard
ARC-AGI 79.5% #47 of 92 Official leaderboard
SimpleBench 50% #54 of 100 Official leaderboard
FrontierMath Tiers 1-3 v2 33.3% #68 of 89 Epoch AI
FrontierMath Tier 4 v2 4.88% #63 of 71 Epoch AI
ProofBench 2.00% #57 of 59 Official leaderboard
ProofBench v1.1 0% #50 of 51 Official leaderboard

Coding

BenchmarkScoreRankSource
SWE-bench Verified 77.6% #32 of 150 Model card
ALE-Bench 946 #56 of 121 Epoch AI
WebDev Arena 1,410 #64 of 105 Epoch AI
Code Arena 1,415 #77 of 124 Official leaderboard
SWE-Bench Pro 54.3% #50 of 77 Model card
Vibe Code Bench 19.2% #86 of 115 Official leaderboard
Code Migration 11.8% #63 of 78 Official leaderboard
WeirdML 32.3% #97 of 112 Official leaderboard
SWE-Bench Pro V2 Hard 56.9% #14 of 16 Official leaderboard
FrontierCode 14% #42 of 46 Epoch AI
SWE-Bench Pro V2 89.9% #15 of 16 Official leaderboard
IOI v2 (Vals) 14.9% #43 of 45 Official leaderboard
Vibe Code Bench 1-100 7.33% #25 of 26 Official leaderboard
Code Migration v1.1 8.82% #29 of 29 Official leaderboard
WeirdML v3 3.90% #25 of 25 Official leaderboard
FrontierSWE 4.12% #21 of 21 Epoch AI

Agents and tools

BenchmarkScoreRankSource
BrowseComp 77.1% #36 of 77 Model card
MCP Atlas 76% #17 of 34 Official leaderboard
τ³-bench Banking 25% #18 of 25 Official leaderboard
Chess Puzzles 21% #60 of 83 Epoch AI
Terminal-Bench 2.1 63.8% #40 of 53 Model card
Toolathlon-Verified 45.5% #22 of 25 Official leaderboard
Agent Arena -0.16 #51 of 53 Official leaderboard

Professional work

BenchmarkScoreRankSource
CorpFin v2 68.6% #7 of 119 Official leaderboard
TaxEval v2 75.3% #20 of 129 Official leaderboard
MedScribe 85.4% #27 of 98 Official leaderboard
PolicyBench 87.7% #15 of 50 Official leaderboard
LegalBench 83% #64 of 138 Official leaderboard
MortgageTax 64% #47 of 86 Official leaderboard
MedCode 41.2% #53 of 96 Official leaderboard
Harvey LAB (Legal Agent Benchmark) 2.08% #49 of 80 Official leaderboard
Legal Research Bench 28.4% #57 of 88 Official leaderboard
Vals Finance Agent v2 46.6% #54 of 80 Official leaderboard
Public Benefits Bench 58.5% #36 of 50 Official leaderboard
Tax Agent Bench 15.3% #58 of 79 Official leaderboard
GDPval-AA v2 1,238 #20 of 26 Model card
$OneMillion-Bench 22.1% #10 of 13 Official leaderboard
Excel Modeling Benchmark 38.4% #61 of 76 Official leaderboard
JobBench 31.1% #13 of 16 Official leaderboard
APEX-Agents 33.8% #48 of 56 Epoch AI
Harvey LAB v1.1 (HLAB-Vals-1.1) 3.75% #30 of 33 Official leaderboard
Vals Finance Agent v2.1 49.9% #33 of 35 Official leaderboard
Vals Index v2.2 29% #31 of 32 Official leaderboard
Vals Finance Index 38.3% #34 of 35 Official leaderboard
Excel Modeling Benchmark v1.1 41% #36 of 37 Official leaderboard
Vals Legal Index 54.2% #33 of 33 Official leaderboard
CAD Arena 0.12 #12 of 12 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
Audio MC 56.6% #3 of 16 Official leaderboard
VoiceBench 91.4% #3 of 10 Model card
CharXiv 82% #22 of 67 Model card
MMAU 77.2% #5 of 14 Model card
MMMU-Pro 73.5% #39 of 72 Model card

Safety and security

BenchmarkScoreRankSource
StrongREJECT 98.6% #1 of 22 Model card
DistressBench 78.5% #6 of 42 Official leaderboard
Agentic BAIM-LLM Evaluation (ABLE) 59.9% #7 of 16 Official leaderboard
FORTRESS 25% #33 of 66 Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2) 50.9% #41 of 71 Official leaderboard
CyberBench v1.1 (Vals) 66.8% #26 of 43 Official leaderboard
CyberBench (Vals) 64.1% #17 of 26 Official leaderboard
DeepsecBench 6.27 #26 of 28 Official leaderboard
SRE-Bench 0% #32 of 33 Official leaderboard
CWE-bench 37% #12 of 12 Official leaderboard

Other

BenchmarkScoreRankSource
ACCoRD 71.2 #43 of 150 Official leaderboard
Conceptual Reasoning Index 52.6 #59 of 149 Official leaderboard
LMCA 37.6% #61 of 139 Epoch AI
SAGE (Vals) 36.5% #63 of 81 Official leaderboard
Vals Multimodal Index 49.4% #27 of 33 Official leaderboard

Thinking Machines's models over time

Thinking Machines's models in color, against every other model on the Epoch Capabilities Index, by release date.

50 75 100 125 150 175 2023 2024 2025 2026 2027 Inkling 50 75 100 125 150 175 2023 2024 2025 2026 2027 Inkling

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Introducing Inkling (Thinking Machines news) 15 Jul 2026

Its system card

Every result in the longer documents the lab published with it.