SSyncDev

Inkling 2 Mini

SyncDev model directory · Profile updated Nov 21, 2026

Compare with another model

Inkling 2 Mini is Thinking Machines' Apache 2.0 looped MoE model (26.4B total, 3.4B active) for agentic coding that runs on a 32 GB laptop.

Key specifications

Specification score
61.6
Context window
1M
License
Apache-2.0

The specification score summarizes recorded features and limits. It is not a benchmark result.

Pricing conditions: Open weights (Apache 2.0). Hosted on Tinker; publisher lists no per-token price on the announcement.

Inputs and outputs

Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.

Accepts

  • Text input
  • Image input

Produces

  • Text output

Capabilities

ReasoningFunction callingVisionOpen weights

What is Inkling 2 Mini?

Can a model that fits in about 20 GB do serious agentic coding? Inkling 2 Mini, released by Thinking Machines Lab on 20 November 2026, is a direct attempt at a yes. It is a 26.4B-parameter mixture-of-experts model with 3.4B active, it runs offline on a 32 GB Mac, and it ships under Apache 2.0.

The unusual part is how it buys extra reasoning without extra size.

Running the middle twice more

Inkling 2 Mini is a looped transformer. Its 24 layers are grouped into four blocks of six. The first and last blocks run once; the two middle blocks run three times per token with shared weights. Stored as 24 layers, it executes 48 layer passes. Each pass through the loop adds about 1.0B effective active parameters, so the true figure is 5.4B per token, not 3.4B.

The intuition is that middle layers do the abstract work, while the outer layers translate to and from embeddings. In the lab's own experiments, looping the middle twelve layers beat looping the last twelve on 27 of 28 evaluations.

Two consequences matter for anyone deploying it. Per-token compute is higher than the active count suggests. And setting reasoning effort to its minimum does not fully switch reasoning off.

Spec Value
Released 20 Nov 2026, open weights
Parameters 26.4B total, 3.4B active, 5.4B effective with looping
Experts 8 of 128 per token
Context Up to 1M tokens
Input / output Text and images in, text out
Precision Trained and served mainly in NVFP4; last four layers MXFP8
Memory About 20 GB
License Apache 2.0

Where it is strong, by the lab's numbers

These scores are publisher-reported. On DeepSWE v1.1 it reaches 48.2%, and on Vals Index v2.2 it posts 46.5%, which the lab says leads every open model under 100B total parameters. SWE-bench Pro v2 Hard lands at 77.9% and SWE-bench Verified at 79.2%. Agentic search looks good too: 79.0% on BrowseComp and 82.7% on MCP Atlas.

Looping does not add memory for facts. Knowledge benchmarks sit where you would expect for its size, and the model leans on search and tools to compensate. It also trails Qwen3.8-27B on multi-turn chat (50.0% on MultiChallenge) and on prompts with many constraints.

Practical notes before you pull the weights

  • On a Mac, Ollama runs it with 32 GB of RAM or more; NVIDIA's DGX Spark handles the native NVFP4 build well.
  • GGUF builds exist for llama.cpp; SGLang, vLLM, NVIDIA Dynamo and TokenSpeed support server inference.
  • Hosted inference and fine-tuning are on Tinker, with a chat playground.
  • Cap output tokens and agent turns. The model can get stuck re-checking its own work, especially while debugging.
  • Expect weaker output outside English; Chinese tests showed clipped phrasing and mixed languages.

Local runtime support for looped models is new, so measured speed varies a lot by runtime. Benchmark on the machine you will actually use.

Picking between the Inkling models

Its bigger sibling, Inkling 2 Small, is the choice when quality ceiling matters more than footprint. The Mini is for laptops, edge boxes and high-throughput agent loops. The simplest trial: give both the same ten repository bug-fix tasks, with turn limits set, and compare pass rate against wall-clock time. A research variable-loop variant was released alongside it for studying how loop count affects chain-of-thought legibility.

See the full Thinking Machines lineup or other code generation models.

How to evaluate Inkling 2 Mini

Compare the model on the work you need to complete.

Tasks and inputs

Inkling 2 Mini is listed for Image-text-to-text, Text generation, Coding & Code Generation. It accepts text, image and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.

Capacity and cost

Documented limits: 1M context. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.

Choosing a deployment

The weights are downloadable under Apache-2.0. Check runtime support and size the complete model and working context for your expected concurrency.

Frequently asked questions

Its middle 12 of 24 layers run three times per token with shared weights, giving 48 layer passes. That raises effective active parameters from 3.4B to about 5.4B per token.

These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.

radixark

DaisyRose-27B

DaisyRose-27B is RadixArk's Apache 2.0, RL-only post-train of Qwen3.8-27B for agentic coding, with a 256K context and image input.

262.14K context

Microsoft

MAI Code 2

MAI Code 2 is Microsoft AI's coding model for multi-file work, debugging and frontend builds, live in GitHub Copilot at $0.50/$2.50 per 1M (promo).

— context

Microsoft

MAI Code 2 Flash

MAI Code 2 Flash is Microsoft AI's low-cost coding model for edits, fixes and explanations, on every GitHub Copilot plan including Free and Student.

— context

Microsoft

MAI Cyber 2 Flash

MAI Cyber 2 Flash is Microsoft AI's security model for vulnerability discovery and threat investigation, with a 1M-token context, shipping in MDASH.

1M context

Google

Gemini 4 Hydrogen

Gemini 4 Hydrogen is Google's fast, lower-cost Gemini 4 model for everyday coding and agentic execution, with a 1M context, xhigh thinking and a free tier.

1.05M context

OpenAI

GPT-6 Cyber

GPT-6 Cyber is OpenAI's restricted security model for authorised vulnerability investigation, reproduction and fix verification, for Daybreak Red customers.

1.05M context