# Inkling

Thinking Machines Lab, released 2026-07-15

**Verdict: conditional.** Inkling is the strongest reason yet to take open-weight multimodal AI seriously. It delivers competitive reasoning and coding with best-in-class open-weight safety and controllable thinking effort, all under Apache 2.0. But Thinking Machines is honest that it's not the strongest overall — factuality lags (SimpleQA 43.9%) and self-hosting demands serious hardware (2TB VRAM). Conditional: a compelling base for organizations that need to own and fine-tune a multimodal model, not for teams wanting peak off-the-shelf performance. Last verified 2026-07-16.

## Pricing

- Input: $1.87 / 1M tokens
- Output: $4.68 / 1M tokens
- Context: Up to 1M tokens native; Tinker 64K and 256K tiers; rates at 50% limited-time discount (list $3.74 / $9.36)

## Strengths

- Open-weight (Apache 2.0) multimodal base — full weights downloadable, modifiable, and deployable on your own infrastructure
- Safety-conscious deployments — best open-weight FORTRESS scores (78.0% adversarial, 95.9% benign refusal)
- Token-efficient reasoning via controllable thinking effort — uses 1/3 the tokens of Nemotron 3 Ultra for equal Terminal Bench performance
- Domain-specific fine-tuning on Tinker — proven uplift in financial reasoning with Bridgewater Associates
- Native multimodal input (text, image, audio) — rare among open-weight models at this capability level
- Pure factuality and knowledge recall tasks — SimpleQA Verified only 43.9%, well behind closed models at 70%+
- Budget-constrained self-hosting — BF16 checkpoint requires ~2TB aggregate VRAM (8x B300 or 16x H200)
- Maximum raw coding leaderboard performance — SWE-bench Pro 54.3% and Terminal Bench 63.8% trail closed frontier

## Benchmarks

- AIME 2026: 97.1% (https://thinkingmachines.ai/news/introducing-inkling/)
- GPQA Diamond: 87.2% (https://thinkingmachines.ai/news/introducing-inkling/)
- HLE (text only): 29.7% (https://thinkingmachines.ai/news/introducing-inkling/)
- SWE-bench Verified: 77.6% (https://thinkingmachines.ai/news/introducing-inkling/)
- Terminal Bench 2.1: 63.8% (https://thinkingmachines.ai/news/introducing-inkling/)
- MCP Atlas: 74.1% (https://thinkingmachines.ai/news/introducing-inkling/)
- MMMU Pro (Standard 10): 73.5% (https://thinkingmachines.ai/news/introducing-inkling/)
- IFBench: 79.8% (https://thinkingmachines.ai/news/introducing-inkling/)
- SimpleQA Verified: 43.9% (https://thinkingmachines.ai/news/introducing-inkling/)

## Sources

- [Tinker model lineup and pricing](https://tinker-docs.thinkingmachines.ai/tinker/models/)
- [TechCrunch — Thinking Machines amps up its bet against one-size-fits-all AI](https://techcrunch.com/2026/07/15/thinking-machines-amps-up-its-bet-against-one-size-fits-all-ai-with-its-first-open-model-inkling/)
- [BenchLM — Thinking Machines Chose Open Weights First](https://benchlm.ai/blog/posts/thinking-machines-chose-open-weights-first)
- [Thinking Machines Lab — Introducing Inkling (official)](https://thinkingmachines.ai/news/introducing-inkling/)
- [WIRED — Thinking Machines Lab Drops Its First Model](https://www.wired.com/story/thinking-machines-lab-releases-its-first-model-inkling/)

Canonical: https://neomanex.com/models/inkling