T
Thinking Machines Lab
San Francisco, CA-based AI company founded 2025. Private. Operates in AI/ML infrastructure. Serves researchers and developers with Tinker, an API for distributed LLM fine-tuning using LoRA, SFT, RL, and distillation.
Plans tracked
42
1 add-on
Pricing changes
2
Latest Nov 9, 2026
Pricing model
Usage-based
No free tier
Pricing Change History
2 changes20262 changes
Week of Nov 2, 2026
Packaging
WeeklyDetected Nov 8, 2026, 8:35 PM ET
Thinking Machines Lab adds Inkling 2 Small at $0.30 and Kimi-K3.1-Pro; Inkling 50% promo prices made permanent
Added Inkling 2 Small ($0.30 prefill/$1.00 sample), Kimi-K3.1-Pro 256K ($6/$25) and serverless Inkling 2 Small; Limited-time 50% discount badges removed from Inkling models, with the discounted prices kept as the new list prices.
Week of Jul 13, 2026
Improvement
WeeklyDetected Jul 20, 2026
Thinking Machines Lab: Inkling models added, Kimi-K2.5 retired
Added Inkling (marked NEW) and Inkling-Small (marked COMING SOON) to Tinker’s supported models list; removed Kimi-K2.5, which had been flagged ‘Retiring July 12’ in the prior version.
Current plans
Checked
| Plan | Price | |
|---|---|---|
| Inkling 2 Small (64K) | Usage-based | |
Also priced by usage
| ||
| GLM-5.3-Flash (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3.5-Super-VL-120B-A12B-BF16 (256K) | Usage-based | |
Also priced by usage
| ||
| GLM-5.3 (256K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-397B-A17B (64K) | Usage-based | |
Also priced by usage
| ||
| DeepSeek-V4.1-Flash (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3.5-Lightning-30B-A3B-BF16 (256K) | Usage-based | |
Also priced by usage
| ||
| Kimi K2.6 (128K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-4B (256K) | Usage-based | |
Also priced by usage
| ||
| Inkling 2 Small (256K) | Usage-based | |
Also priced by usage
| ||
| Inkling (64K) | Usage-based | |
Also priced by usage
| ||
| Inkling (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3-Ultra-550B-A55B-BF16 (64K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3-Ultra-550B-A55B-BF16 (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3-Super-120B-A12B-BF16 (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3-Nano-30B-A3B-BF16 (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.8-27B (256K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.6-35B-A3B (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.6-27B (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3-8B (32K) | Usage-based | |
Also priced by usage
| ||
| GPT-OSS-120B (32K) | Usage-based | |
Also priced by usage
| ||
| GPT-OSS-20B (32K) | Usage-based | |
Also priced by usage
| ||
| DeepSeek-V4.1-Flash-1017 (256K) | Usage-based | |
Also priced by usage
| ||
| Inkling 2 Small Serverless Inference (256K) | Usage-based | |
Also priced by usage
| ||
| Inkling-Small Serverless Inference (256K) | Usage-based | |
Also priced by usage
| ||
| Inkling-Small (64K) | Usage-based | |
Also priced by usage
| ||
| Inkling-Small (256K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3.5-Lightning-30B-A3B-BF16 (64K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3.5-Super-VL-120B-A12B-BF16 (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.8-27B (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-9B-Base (64K) | Usage-based | |
Also priced by usage
| ||
| Nemotron-3-Super-120B-A12B-BF16 (64K) | Usage-based | |
Also priced by usage
| ||
| Kimi-K3.1-Pro (256K) | Usage-based | |
Also priced by usage
| ||
| Kimi K2.6 (32K) | Usage-based | |
Also priced by usage
| ||
| GPT-OSS-120B (128K) | Usage-based | |
Also priced by usage
| ||
| DeepSeek-V3.1 (32K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.6-35B-A3B (256K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-397B-A17B (256K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-35B-A3B-Base (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-9B (64K) | Usage-based | |
Also priced by usage
| ||
| Qwen3.5-4B (64K) | Usage-based | |
Also priced by usage
| ||
| Inkling Serverless Inference (256K) | Usage-based | |
Also priced by usage
| ||
| StorageAdd-on | $0.10 per GB per month | |
Company Facts
CategoryAI & Machine Learning
Free tierNo
Employees11-50
Jun 24, 2026
Last checked
Websitethinkingmachines.ai
How Thinking Machines Lab compares with its peers
| Company | Entry price |
|---|---|
O OpenAILast change: Nov 13, 2026 | $8 /mo |
A AnthropicLast change: Nov 11, 2026 | $17 /mobilled annually |
G GoogleLast change: Oct 12, 2026 | Contact |
M MicrosoftLast change: Oct 26, 2026 | $5 /user/mobilled annually |
H HuggingfaceLast change: Nov 2, 2026 | $9 /mo |
Thinking Machines Lab pricing FAQ
- How much does Thinking Machines Lab cost?
- Thinking Machines Lab prices on usage: the Inkling 2 Small (64K) plan is $0.30 per million prefill tokens + $0.03 per million cached prefill tokens.
- Does Thinking Machines Lab have a free plan?
- No. Thinking Machines Lab does not list a free tier.
- What pricing model does Thinking Machines Lab use?
- Thinking Machines Lab uses usage-based pricing across 42 plans.
- When did Thinking Machines Lab last change its pricing?
- Nov 9, 2026: Thinking Machines Lab adds Inkling 2 Small at $0.30 and Kimi-K3.1-Pro; Inkling 50% promo prices made permanent.
- How often has Thinking Machines Lab changed its pricing?
- Pulse has recorded 2 pricing changes for Thinking Machines Lab since it started tracking on Jun 24, 2026; the pricing page was last checked on Nov 9, 2026.
- What does “tracked since” mean?
- It's the date of the first saved copy of the pricing page that Pulse checked for changes against a later copy. For a few companies we hold older copies that Pulse never checked for changes, and those don't change this date. If Pulse has never checked two copies, it's the date of the first saved copy.
- Who are Thinking Machines Lab's pricing peers?
- Pulse compares Thinking Machines Lab with OpenAI, Anthropic, Google, Microsoft and Huggingface.