Inkling-Small

Inkling-Small places in the top three on 2 benchmarks.

Thinking Machines Released 15 Jul 2026

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Thinking Machines Jul 2026

Inkling-Small

Open weights

150.5 #59 of 271

Epoch Capabilities Index

IFBench 82.2% #3 /70
MMMU-Pro 74% #35 /72
SciCode 48.7% #12 /47

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

Benchmark	Score	Rank	Source
IFBench	82.2%	#3 of 70	Model card
Humanity's Last Exam	47.8%	#45 of 147	Model card
LMArena (Elo)	1,406	#123 of 333	Official leaderboard
SEA-HELM	71.5	#28 of 62	Official leaderboard
Global MMLU Lite 3.0	86.7%	#2 of 3	Model card
Creative Writing v3	58.9	#85 of 120	Official leaderboard
SimpleQA Verified	19.1%	#74 of 84	Epoch AI

Math and science

Benchmark	Score	Rank	Source
GPQA Diamond	88.5%	#52 of 219	Epoch AI
AIME 2026	95.5%	#9 of 34	Model card
OTIS Mock AIME 2024-2025	90%	#60 of 202	Epoch AI
CritPt	8.29%	#63 of 160	Epoch AI
ARC-AGI-2	40.1%	#44 of 95	Official leaderboard
ARC-AGI	84%	#44 of 92	Official leaderboard
HMMT 2026	90.2%	#10 of 18	Model card
FrontierMath Tiers 1-3 v2	46.3%	#56 of 89	Epoch AI
FrontierMath Tier 4 v2	17.1%	#53 of 71	Epoch AI
ProofBench	5.00%	#52 of 59	Official leaderboard
ProofBench v1.1	6.00%	#48 of 51	Official leaderboard
Prophet Arena	0.12	#1 of 1	Model card

Coding

Benchmark	Score	Rank	Source
SWE-bench Verified	80.2%	#18 of 150	Model card
SciCode	48.7%	#12 of 47	Model card
SWE-Bench Pro	55.9%	#44 of 77	Model card
WebDev Arena	1,407	#65 of 105	Epoch AI
Code Arena	1,409	#78 of 124	Official leaderboard
Code Migration	13.7%	#59 of 78	Official leaderboard
Vibe Code Bench	19.1%	#87 of 115	Official leaderboard
Image-to-WebDev Arena	1,410	#50 of 55	Official leaderboard
FrontierCode 2.0	3.21%	#24 of 26	Official leaderboard
Code Migration v1.1	10.4%	#28 of 29	Official leaderboard
IOI v2 (Vals)	9.33%	#44 of 45	Official leaderboard

Agents and tools

Benchmark	Score	Rank	Source
MCP Atlas	79.2%	#12 of 34	Official leaderboard
BrowseComp	77.4%	#33 of 77	Model card
Terminal-Bench 2.1	64.7%	#38 of 53	Model card
Toolathlon-Verified	54.4%	#18 of 25	Official leaderboard
τ³-bench Banking	15.5%	#16 of 20	Model card
Mystery Game Puzzles	6.00%	#76 of 80	Epoch AI
Agent Arena	-0.17	#52 of 53	Official leaderboard

Professional work

Benchmark	Score	Rank	Source
CorpFin v2	69.6%	#6 of 119	Official leaderboard
TaxEval v2	75.5%	#18 of 129	Official leaderboard
MedScribe	84.1%	#37 of 98	Official leaderboard
LegalBench	83%	#62 of 138	Official leaderboard
MortgageTax	62.3%	#52 of 86	Official leaderboard
Harvey LAB (Legal Agent Benchmark)	1.67%	#53 of 80	Official leaderboard
Public Benefits Bench	59.4%	#34 of 50	Official leaderboard
Legal Research Bench	25.5%	#61 of 88	Official leaderboard
MedCode	37.9%	#69 of 96	Official leaderboard
Tax Agent Bench	15.4%	#57 of 79	Official leaderboard
GDPval-AA v2	1,269	#19 of 26	Model card
Vals Finance Agent v2	41.3%	#62 of 80	Official leaderboard
Excel Modeling Benchmark	31.8%	#64 of 76	Official leaderboard
CAD Arena	0.14	#11 of 12	Official leaderboard
Vals Legal Index	57.8%	#31 of 33	Official leaderboard
Harvey LAB v1.1 (HLAB-Vals-1.1)	2.92%	#32 of 33	Official leaderboard
Excel Modeling Benchmark v1.1	40.8%	#37 of 37	Official leaderboard
Vals Finance Agent v2.1	44.8%	#35 of 35	Official leaderboard
Vals Finance Index	36.3%	#35 of 35	Official leaderboard
Vals Index v2.2	27.6%	#32 of 32	Official leaderboard
AA-Briefcase	917	#6 of 6	Model card

Vision, audio and robotics

Benchmark	Score	Rank	Source
Audio MC	54.9%	#4 of 16	Official leaderboard
CharXiv	81.3%	#26 of 67	Model card
VoiceBench	90.1%	#4 of 10	Model card
MMAU	77%	#6 of 14	Model card
MMMU-Pro	74%	#35 of 72	Model card
Vision Arena	1,209	#73 of 136	Official leaderboard

Safety and security

Benchmark	Score	Rank	Source
StrongREJECT	98.4%	#2 of 22	Model card
CyberBench (Vals)	68.8%	#14 of 26	Official leaderboard
CyberBench v1.1 (Vals)	64.1%	#30 of 43	Official leaderboard
FORTRESS	71.6%	#5 of 6	Model card

Other

Benchmark	Score	Rank	Source
Vals Multimodal Index	50%	#26 of 33	Official leaderboard
SAGE (Vals)	32.7%	#70 of 81	Official leaderboard
Artificial Analysis Intelligence Index v4.1	40.0	#7 of 7	Model card

Thinking Machines's models over time

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

Its launch post

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

Inkling-Small (Thinking Machines news) 30 Jul 2026

AA Index v4.1 40
SWEBench Verified (bash-only harness) 80.2
SWEBench Pro public 55.9
Terminal Bench 2.1 (best harness) 64.7
SciCode 48.7
GDPval-AA v2 1269
MCP Atlas public 79.6 / all 79.2 79.6
Tau 3 Banking 15.5
BrowseComp with context management 77.4
Toolathlon Verified 54.4
AA-Briefcase 917
GPQA Diamond 89.5
Best of HLE text only 31.6 / HLE with tools 47.8 47.8
AIME 2026 95.5
HMMT Feb 2026 90.2
CritPt 8.3
ARC-AGI-1 84
ARC-AGI-2 40.1
SimpleQA Verified 20.6
AA Omniscience index -9
IFBench 82.2
Global-MMLU-Lite 86.7
StrongREJECT 98.4
FORTRESS adversarial 71.6 (benign 96.9) 71.6
MMMU Pro Standard 10 74
Best of Charxiv RQ original 77.4 / with python 81.3 81.3
Audio MC 54.9
MMAU 77
VoiceBench 90.1
Best of ForecastBench Brier Index no search 61.3 / with search 61.5 61.5
Prophet Arena Brier Score (lower is better) 0.124

Its system card

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

Inkling-Small Model Card 30 Jul 2026