Signal SentrySearchUpdated Nov 20, 05:30 UTCAM Drop
Models · scorecard

Inkling 2 Small

What the vendor claims, next to what independent boards measured. The two are never mixed in one column.

At a glance

Maker
Thinking Machines
Price per 1M tokens
$0.3 in · $1 out
Context window
1.05M
Input → output
text, image → text
Added to OpenRouter
Nov 9, 2026

Source: OpenRouter, as of Nov 20, 06:00 UTC. Model page on OpenRouter

What the vendor claims

Quoted word for word from the vendor's own launch post or model card. Claims, not measurements.

No vendor claims recorded yet. They come from launch posts and model cards the Drops covered.

Independent results

Measured by independent sources (Epoch AI, LMArena, OpenRouter's benchmark runs, τ²-bench, Vectara), not by the vendor. "Listed as" is the source's own name for the model, so you can check our match.

Independent benchmark results for this model, each with its source and date
BoardResultRank
LMArena · WebDevListed as: Inkling 2 SmallLMArena (CC BY 4.0), as of Nov 18, 2026. Source1,644Arena score27
LMArena · AgentListed as: Inkling 2 SmallLMArena (CC BY 4.0), as of Nov 18, 2026. Source0.0151Score26
OpenRouter · GPQA DiamondListed as: Inkling 2 Smallvia Nebius, the most accurate providerSource: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source91.4%Accuracy26
OpenRouter · GPQA Diamond: cost per taskListed as: Inkling 2 Smallvia Nebius, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source$0.031Cost per task20
OpenRouter · GPQA Diamond: time per taskListed as: Inkling 2 Smallvia Nebius, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source320 sTime per task46

Price history

OpenRouter's list price per 1M tokens in each of our snapshots.

  • Output now $1
  • Input now $0.3

Unchanged across our 15 snapshots.

Show the numbers
This model's OpenRouter price in each of our snapshots
SnapshotInput $/1MOutput $/1M
Nov 20, 06:04 UTC$0.3$1
Nov 19, 16:04 UTC$0.3$1
Nov 19, 04:04 UTC$0.3$1
Nov 18, 18:04 UTC$0.3$1
Nov 18, 04:03 UTC$0.3$1
Nov 17, 16:01 UTC$0.3$1
Nov 17, 04:04 UTC$0.3$1
Nov 16, 19:54 UTC$0.3$1
Nov 16, 04:05 UTC$0.3$1
Nov 15, 16:04 UTC$0.3$1
Nov 15, 04:03 UTC$0.3$1
Nov 14, 16:04 UTC$0.3$1
Nov 14, 04:05 UTC$0.3$1
Nov 13, 16:05 UTC$0.3$1
Nov 13, 04:04 UTC$0.3$1

Source: OpenRouter, as of Nov 20, 06:04 UTC

In the Drops

Models

Thinking Machines releases Inkling 2 Small on Tinker; Apache 2.0 weights planned

Thinking Machines says Inkling 2 Small is a 594B-parameter mixture-of-experts model with 17B active and image input. It is on Tinker now with closed weights; Apache 2.0 weights are planned next week after outside safety testing.

Why it matters: A US lab plans permissive open weights for a model it pitches for agentic coding. It also says early checkpoints faked results, so read its alignment notes before fine-tuning.

From Drop #069 · Fri, Nov 6, 2026 · Morning