SI Index Data as of Oct 9, 2026

Inkling Small: Price, Context, Benchmarks, and Release Details

thinkingmachines provisional listing open weights · first seen Oct 8, 2026 · score computed Oct 9, 2026, 01:50 UTC
53.6
SI Score (method si-v3-retained-evidence-2)
#94 of 140 ranked
100% confidence 100 percent, Full confidence — 3 of 3 expected sources in
Coverage 100% of expected source weight · 100% confidence at 80% coverage

Pillar breakdown

Coding (weight 40 percent) —
Math (weight 15 percent) 43.4
Preference (weight 15 percent) 74.4
Reasoning (weight 30 percent) 53.1

Pillar weights: reasoning 30% · math 15% · coding 40% · preference 15%. Benchmark results use fixed 0–100 scales before averaging; incomplete evidence is shrunk toward 50.

Facts

Input price / 1M
not yet reported
Output price / 1M
not yet reported
Context window
1Mmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Max output
1Mmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Released
Jul 30, 2026models.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Open weights
Yesmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
License
Apache-2.0models.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Input modalities
text, image, audiomodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗

Benchmark results

21 results
Benchmark Raw result Normalized (0–100) Pillar
arc agi v1 public eval 86.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] thinky-inkling-small-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		86.6 reasoning
arc agi v1 public eval 66.1%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] thinky-inkling-small-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		66.1 reasoning
arc agi v1 public eval 77.8%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] thinky-inkling-small-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		77.8 reasoning
arc agi v1 public eval 89.4%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] thinky-inkling-small-xhighPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		89.4 reasoning
arc agi v1 semi private 78%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] thinky-inkling-small-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		78.0 reasoning
arc agi v1 semi private 52.5%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] thinky-inkling-small-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		52.5 reasoning
arc agi v1 semi private 67%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] thinky-inkling-small-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		67.0 reasoning
arc agi v1 semi private 84%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] thinky-inkling-small-xhighPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		84.0 reasoning
arc agi v2 public eval 25.7%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] thinky-inkling-small-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		25.7 reasoning
arc agi v2 public eval 6.4%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] thinky-inkling-small-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		6.4 reasoning
arc agi v2 public eval 13.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] thinky-inkling-small-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		13.6 reasoning
arc agi v2 public eval 39.3%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] thinky-inkling-small-xhighPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		39.3 reasoning
arc agi v2 semi private 33.1%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] thinky-inkling-small-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		33.1 reasoning
arc agi v2 semi private 4.9%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] thinky-inkling-small-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		4.9 reasoning
arc agi v2 semi private 13.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] thinky-inkling-small-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		13.6 reasoning
arc agi v2 semi private 40.1%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] thinky-inkling-small-xhighPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		40.1 reasoning
frontiermath tier 4 v2 17.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		17.1 math
frontiermath tiers 1 3 v2 46.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		46.3 math
gpqa diamond 88.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		88.5 reasoning
lmarena text 1413.7 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 2, 2026 Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗ 		74.4 preference
otis mock aime 2024 2025 90%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		90.0 math

Normalization uses fixed absolute 0–100 scales for each unit, independently of other models that have a result on each benchmark. Coverage and evidence breadth still affect the composite; compare the evaluation conditions before reading a small score gap as decisive. “lab-reported” marks the provider's own published figure.

Sources: in and pending

in Reported (3)

  • ARC Prize · arrived Oct 8, 2026
  • Epoch AI Benchmarking · arrived Oct 8, 2026
  • LMArena / Arena · arrived Oct 8, 2026

pending Awaiting (0)

Every expected source has reported for this model.

The confidence % rises as pending sources publish. Some sources never cover some models — that is why 100% confidence arrives at 80% of expected weight, not at full coverage.