Inkling 2 Small Ships on Tinker With Weights Closed | Zeniteq
Loading
Loading article
Zeniteq is now available on mobile for iPhone, iPad and Apple Watch.Download nowLearn more
Inkling 2 Small Ships on Tinker With Weights Closed
Thinking Machines’ new model is available for hosted sampling and fine-tuning, while downloadable Apache 2.0 weights remain a conditional plan for next week.
Thinking Machines Lab released Inkling 2 Small on November 5, 2026, offering immediate access to sampling and fine-tuning via its Tinker platform. The model utilizes a Mixture-of-Experts architecture with 594 billion total parameters but only 17 billion active parameters, necessitating significant hardware for self-hosting once weights are released. While the company promises an Apache 2.0 release next week following safety testing, developers currently cannot download weights or export fine-tuning checkpoints. Builders should evaluate the model's agentic coding capabilities in the Tinker Playground now, but defer long-term deployment plans until the full model becomes portable.
Thinking Machines Lab released Inkling 2 Small through Tinker on November 5, 2026, giving developers immediate access to sampling, fine-tuning and the Tinker Playground. The second-generation model has 594 billion total parameters, 17 billion active parameters, image-input support and adjustable reasoning effort, according to the company’s announcement.
The weights are still closed. Thinking Machines intends to release them under Apache 2.0 the following week, after external safety testing and red-teaming are complete. Downloadable fine-tuning checkpoints are scheduled to arrive alongside that release.
That distinction matters for builders: hosted experimentation can begin now, but self-hosting and checkpoint export still depend on a future release.
What Builders Can Use in Tinker Now
Thinking Machines lists three available ways to work with Inkling 2 Small: try it in Tinker Playground, sample it through Tinker, or fine-tune it through Tinker. Sampling means requesting model outputs; fine-tuning means adapting the model using additional training data.
The current and planned access paths break down like this:
Capability
Status in the November 5 announcement
Try the model in Tinker Playground
Available
Sample outputs through Tinker
Available
Fine-tune through Tinker
Available
Download the full model weights
Planned for the following week, after safety testing
For a team considering customization, the practical split is between starting training and taking the resulting checkpoint elsewhere. Tinker supports the former now. The announcement ties downloadable checkpoints to the later weight release.
I’d use this window to evaluate task quality before committing to a deployment plan. Builders can examine whether the model follows their instructions, handles their images and completes their workflows. Decisions that require portable checkpoints should wait for the actual downloads and release documentation.
Why 17B Active Does Not Make This a 17B Deployment
Inkling 2 Small uses a Mixture-of-Experts, or MoE, architecture. In a sparse MoE, a token uses a subset of the model’s parameters instead of activating the entire model for every computation. Thinking Machines reports 594B total parameters and 17B active parameters, where “B” means billion. It also reports support for image inputs and variable reasoning effort. (Announcement)
The active count is about 2.9% of the total parameter count. That helps explain the appeal of the design: the model can have a large overall parameter pool while using a much smaller portion during processing.
It does not follow that the eventual release will fit wherever a dense 17B model fits.
A simple storage calculation illustrates the difference. If all 594 billion parameters were stored at two bytes each, the weights alone would occupy approximately 1.188 terabytes, using decimal units. That is arithmetic based on the published parameter count, not a disclosed Inkling checkpoint format or hardware requirement.
Actual deployment needs will depend on weight precision, quantization, runtime support, how experts are distributed across devices, and memory used beyond the weights. The announcement does not specify practical self-hosting requirements.
My view is that the total parameter count is the more useful starting point for deployment planning, while the active count helps explain the compute design. Treating “17B active” as a memory specification would skip the largest unresolved infrastructure question.
Variable reasoning effort introduces another choice: how much inference work to spend on a response. The announcement does not provide enough detail to turn that feature into a reliable latency or cost estimate.
Coding and Visual Work Depend on the Agent Loop
Thinking Machines positions Inkling 2 Small for agentic coding, knowledge work, reasoning and long-horizon autonomous tasks. It says the model received reinforcement learning on hundreds of thousands of long-horizon coding tasks across multiple agent harnesses and context-compaction strategies.
The company also says some training tasks used an agentic grader that evaluated the model’s approach, extending the reward beyond whether code passed unit tests. These are descriptions of its training process, not independently verified explanations of the reported gains. (Announcement)
The visual examples make the intended workflow clearer. Thinking Machines shows a procedurally generated Three.js city simulator, a Python PC emulator and Blender-created scenes. It says the model can use visual feedback to build and refine websites, games and interactive 3D designs.
Image input should not be confused with native image generation. The cited examples involve the model working through software and tools. Their practical value depends on the surrounding system: what the agent can execute, what feedback it receives, and how it preserves useful context over many steps.
For a builder, a useful evaluation would therefore include the full loop: make a change, inspect the result, identify a problem and revise it. A strong first answer alone would miss the behavior this release is being positioned to improve.
What the Vendor’s Benchmarks Show
Thinking Machines reports an Artificial Analysis Intelligence Index score of 31, which it describes as a substantial improvement over its previous models. The score appears in the company’s own announcement; the supplied evidence contains no separate independent measurement confirming it.
Its published coding comparisons also show a mixed competitive picture. The following figures are all reported by Thinking Machines:
Inkling trails the listed DeepSeek model on three of these four rows and leads it on PostTrainBench. That last result is not a table-wide win: Thinking Machines lists GLM-5.3-Flash at 36.8% on PostTrainBench.
These results support a narrower reading than a blanket claim of superiority. The company is reporting gains over its own earlier generation and competitive results against other models, with the ordering changing by task.
The Long-Horizon Test Has Important Conditions
Thinking Machines also publishes an adapted PMPP-Hard GPU kernel-optimization evaluation covering 69 tasks. Its stated setup uses four independent attempts, a 10-hour budget, each model’s maximum available reasoning setting, the PI harness and a shared context-compaction policy.
Under those conditions, Thinking Machines reports a mean of 54.25 tasks solved out of 69 for Inkling 2 Small at 10 hours, compared with 55.25 for the DeepSeek V4.1 Flash entry in that evaluation. It also reports a pass@4 count of 64 for Inkling. These are different summaries: the mean describes performance across attempts, while pass@4 captures success across the four opportunities. (Evaluation details)
The shared setup makes this comparison more interpretable than scores without disclosed conditions. It remains a vendor-run evaluation, even though the underlying benchmark is external. A shared time limit also does not establish equal inference cost, and the announcement does not supply pricing that would settle that comparison.
What Must Happen Before the Apache 2.0 Release
Thinking Machines states its intention directly: “We intend to make the full weights available under an Apache 2.0 license next week”. The same sentence makes completion of external safety testing and red-teaming part of that plan. (Announcement)
“Next week” refers to the week following the November 5 announcement. It is a target, not confirmation that a repository or downloadable checkpoint is already available.
Three pieces matter for builders:
Testing must finish. The announcement places external safety testing and red-teaming before the intended release.
The full weights must become downloadable. A stated license plan is not access to the model files.
Fine-tuning checkpoints must become exportable. Thinking Machines schedules those downloads alongside the weights.
The scope is also worth keeping precise. The announcement promises a planned Apache 2.0 release of the full weights. It does not announce an open-source release of Tinker’s hosted service or infrastructure.
If completed as described, the release would give developers another substantial open-weight option and a path from hosted customization toward downloadable checkpoints. Until then, Inkling 2 Small is a shipped hosted model with a conditional open-weight release ahead of it.
Community Discussion Is Already Looking Past Hosted Access
The Hacker News discussion includes comparisons with DeepSeek and debate about the significance of a prospective U.S.-developed open-weight model. Commenter segmondy described the published results favorably, while other participants disputed broader claims about competitive standing and the U.S. open-weight gap.
Those comments document interest and disagreement within one forum thread. They are not representative user testing or independent benchmark evidence. The most consequential questions remain practical: whether the planned downloads arrive, what running them requires, and whether performance holds up in a builder’s own workflow.
Frequently Asked Questions
3 questions
1
Can I Use Inkling 2 Small Now?
Yes. Thinking Machines says it is available in Tinker Playground and through Tinker for sampling and fine-tuning.
2
Are the Weights Open Under Apache 2.0 Yet?
No. In the November 5 announcement, the weights remain closed. Thinking Machines intends to release them the following week after external safety testing and red-teaming.
3
Can I Download My Fine-Tuning Checkpoint?
Not under the availability described in the announcement. Thinking Machines says downloadable checkpoints will arrive alongside the planned weight release.
