Xylolabs

Inkling 2 Mini GGUF

Base model License Code

These llama.cpp GGUFs of Inkling 2 Mini include a separate vision projector and omit the MTP drafter.

Files

File Size
Inkling-2-Mini-NVFP4.gguf 22.92 GB
Inkling-2-Mini-NVFP4-Q8_0.gguf 17.63 GB
Inkling-2-Mini-NVFP4-XL.gguf 17.61 GB
mmproj-Inkling-2-Mini-BF16.gguf 641.81 MB
mmproj-Inkling-2-Mini-Q8_0.gguf 345.91 MB

How to run

Use llama.cpp release b12611 or later, with the Hugging Face CLI to download the files. Earlier builds skip the per-expert NVFP4 scales on the second and third loop passes and give wrong output with these files.

hf download Xylolabs/Inkling-2-Mini-GGUF Inkling-2-Mini-NVFP4-Q8_0.gguf mmproj-Inkling-2-Mini-BF16.gguf --local-dir .
llama-server -m Inkling-2-Mini-NVFP4-Q8_0.gguf --mmproj mmproj-Inkling-2-Mini-BF16.gguf --jinja

For text-only use, omit --mmproj and the projector filename.

Accuracy

Mean KL divergence from the F32 reference (lower is better) and top-1 agreement with it, measured with llama.cpp on 512-token chunks.

Quant Size WikiText-2 KL top-1 Chat KL top-1 Held-out KL top-1
NVFP4 22.92 GB 0.0023 98.0% 0.0219 97.4% 0.0021 98.1%
NVFP4-Q8_0 17.63 GB 0.0030 97.6% 0.0194 97.3% 0.0027 97.9%
NVFP4-XL 17.61 GB 0.0026 97.8% 0.0214 97.2% 0.0024 97.9%
Unsloth UD-Q4_K_XL 17.63 GB 0.0029 97.6% 0.0221 97.2% 0.0027 97.8%
Unsloth UD-Q8_K_XL 22.87 GB 0.0024 97.9% 0.0202 97.4% 0.0022 98.3%
Others 0 Q4_K_M 17.38 GB 0.0167 94.4% 0.0615 94.0% 0.0139 94.8%
Others 0 Q6_K 23.40 GB 0.0046 97.2% 0.0269 96.4% 0.0040 97.5%
Others 5 IQ4_NL 17.63 GB 0.0087 95.9% 0.0368 95.7% 0.0068 96.3%
Others 5 Q8_0 27.82 GB 0.0030 97.7% 0.0215 96.9% 0.0028 98.0%
llama.cpp baseline Q4_K_M 17.48 GB 0.0167 94.4% 0.0594 94.1% 0.0131 95.0%
WikiText-2 mean KL divergence by text GGUF size WikiText-2 top-1 agreement by text GGUF size

Which file for your hardware

File size, plus the KV cache at the stated context (f16), plus about 2 GB of runtime overhead, must fit in 90% of the memory. Its layers 6-17 run three times, each pass with its own cache: 8 caches use full attention (8 KV heads, head size 128) and 40 use a 1,024-token sliding window, a fixed cache of about 0.25 GB.

Hardware Memory Recommended file Context that fits Notes
RTX 4090 / RTX 3090 24 GB Inkling-2-Mini-NVFP4-XL.gguf (17.6 GB) 32K
RTX 5090 32 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 64K
RTX 6000 Ada, or 2 x 24 GB 48 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 512K Two 24 GB cards need the layers split across them.
One H100 80 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M
RTX PRO 6000 Blackwell (GDDR7) 96 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M
DGX Spark 128 GB unified Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M
Mac Studio M5 Max 128 GB unified Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M The GPU may use about 75% of unified memory by default; sudo sysctl iogpu.wired_limit_mb=N raises it.
One H200 (HBM3e) 141 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M
One B200 (about 180 GB usable in a DGX B200) 192 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M
Mac Studio M5 Ultra 256 GB unified Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M The GPU may use about 75% of unified memory by default; sudo sysctl iogpu.wired_limit_mb=N raises it.
Mac Studio M5 Ultra 512 GB unified Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M The GPU may use about 75% of unified memory by default; sudo sysctl iogpu.wired_limit_mb=N raises it.
One B300 (HBM3e; usable a little less) 288 GB Inkling-2-Mini-NVFP4.gguf (22.9 GB) 1M

License and notices

These GGUF editions are licensed under Apache 2.0 and subject to Thinking Machinesโ€™ Model Acceptable Use Policy. The NOTICE identifies the copyright holder and Xylolabsโ€™ changes. Xylolabs is independent of Thinking Machines Lab.

Updates

  • 2026-11-22: Added a hardware recommendation table.
  • 2026-11-22: Added Inkling-2-Mini-NVFP4-XL.gguf.
  • 2026-11-22: llama.cpp release b12611 or newer runs these files; earlier builds give wrong output.
  • 2026-11-21: First release.

Thinking Machines' original model card

Thinking Machines Labโ€™s original model card follows.

Inkling 2 Mini

Playground | Tinker Cookbook

1. General Information

Inkling 2 Mini is a general-purpose multimodal model that accepts text and image inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages. The model is designed to be used by developers building AI-powered applications, including agentic and tool-use systems, coding assistants, chatbots, and retrieval-augmented generation systems, and is suitable for general-purpose conversational use, instruction-following, and other natural language and visual tasks. It is released with open weights to support research, fine-tuning and integration into third-party products by downstream developers.

Languages: English, general multilingual support

Model provider: Thinking Machines Lab, Inc.

Date of release: November 20, 2026

License: Apache 2.0

Model Acceptable Use Policy

2. Getting Started

The weights are available for download through Hugging Face here.

Developers can also access the model using the Tinker fine-tuning platform, which provides API access.

  • For accessing our model via Tinker, you can get started by referring to the Tinker Cookbook here, and installing the tml-renderers package here.
  • Tinker playground also provides inference-only access to the model here.

Required hardware

The model is distributed as a single checkpoint of about 20GB, in mixed NVFP4 and MXFP8 precisions.

Given its small size, it is possible to run this checkpoint on local AI compute for inference. We support:

  • 1x NVIDIA DGX Spark
  • Apple silicon devices with 32GB or more memory (MacBooks, Mac minis and Studios)
  • Other GPUs and platforms are possible, with 32GB or more of VRAM recommended

Required software

Inkling 2 Mini supports local inference using the following frameworks, along with their dependency libraries:

Inkling 2 Mini can also be post-trained with:

3. Model Properties

Model type

Multimodal autoregressive looped transformer

Architecture type

A 24-layer decoder-only transformer with sparse Mixture-of-Experts (MoE) feed-forward layers. The middle 12 layers run three times with shared weights, giving 48 layer applications per token from 24 layers of weights. In each layer, each token is routed to 8 of 128 experts. Attention is a hybrid of local and global layers. The model is natively multimodal โ€“ images are encoded via a vision transformer โ€“ with image and text inputs projected into a shared hidden space and processed jointly by the decoder.

Parameters

26.4B, 3.4B active

Note: each extra pass through the looped layers adds about 1.0B effective active parameters. Since the model runs with three passes, there are 5.4B effective active parameters per token.

Numerics support

NVFP4 and MXFP8

Input modalities

  • Text: UTF-8 encoded text
  • Image: Any pixel-based image input. For optimal performance, each image dimension should be between 40px to 4096px.

Inkling 2 Mini supports a context window of up to 1M tokens.

Output modalities

Text

4. Training

Training data includes a broad variety of content types, including text, images, video. Training data for the model was drawn from publicly available sources, acquired from third-parties, generated internally, or synthetically generated or augmented. Publicly available data includes content from the public internet and publicly accessible repositories.

Data curation includes cleaning, processing, and modifying datasets. These processing steps, which vary by data type, may include deduplication and filtering to remove junk or other low-quality data, or to advance safety or other objectives.

5. Evaluations

Inkling 2 MiniGemma 4 31BMuse GlimmerQwen3.8-27BNemotron 3.5 Super VL
Model Info
Params (B)
activated / total
5/2631/3130/3027/2712/125
Third Party Index
AA Index (v5.0)20.14.46.320.113.6
Vals Index (v2.2)46.5%โ€“โ€“โ€“29.8%
Coding and Autoresearch
FrontierCode v2.0
stirrup v0.2.1
15.1%0.0%1.5%14.9%6.1%
DeepSWE v1.1
mini-swe-agent v2.4.6
48.2%0.3%4.4%42.2%24.7%
Terminal Bench v4.0
mini-swe-agent v2.4.6
5.6%0.0%0.5%5.6%0.0%
Terminal Bench v2.1
Inkling 2 Mini in pi v0.83.0
71.6%43.4%51.7%73.0%60.7%
SWEBench Pro v2 (Hard)
mini-swe-agent v2.4.6
77.9%14.2%44.3%73.8%44.1%
SWEBench Multilingual
Inkling 2 Mini with bash only
73.7%43.0%66.9%70.0%67.7%
SWEBench Verified
bash only
79.2%46.1%76.0%77.8%71.1%
Vibe Code Bench v1.1
openhands (vals)
77.9%โ€“โ€“64.9%3.5%
PMPP-Hard (AUC, 0โ€“10h)
pi v0.83.0
455.82โ€“โ€“262.50โ€“
Agentic Search, Tool Use, and Knowledge Work
BrowseComp
Inkling 2 Mini with compaction
79.0%37.0%63.9%61.6%78.9%
DeepSearchQA
Inkling 2 Mini with compaction
85.3%61.7%74.6%79.7%77.7%
MCP Atlas
Public
82.7%54.2%75.5%77.3%67.7%
AA-Briefcase v1.112633694781391999
GDPVal-AA v2.1131962779214271162
JobBench49.6%8.2%16.1%33.4%36.5%
AutomationBench-AA v1.0.641.8%4.9%6.8%48.2%14.8%
EnterpriseOps-Gym42.0%28.4%34.7%44.2%28.1%
Vals Finance Index50.1%โ€“โ€“โ€“41.0%
AA-AnalystAgent46.3%โ€“20.0%โ€“โ€“
Diligence Bench
Best harness
48.6%โ€“40.9%46.4%45.0%
Vision Interpretation and Creation
GDP.pdf16.8%6.0%11.8%16.4%8.0%
PerceptionBench57.5%40.7%42.8%52.4%36.8%
MMMU Pro
Standard 10
74.9%73.4%74.3%76.3%72.3%
OmniDocBench v1.692.8%82.5%90.4%91.5%77.9%
ScreenSpot Pro
original / with Python
76.5/87.7%5.9%/โ€“75.4%/โ€“57.6/78.9%43.5%/โ€“
Chartography
original / with Python
22.1/72.5%12.0/11.8%17.4/25.1%12.2/52.1%9.7/29.3%
ZeroBench (main)
original / with Python
17.0/34.0%9.0/12.0%9.0/17.0%12.0%/โ€“11.0/12.0%
MuirBench77.5%72.3%77.6%65.9%72.7%
BVB68.4%47.0%50.5%70.7%52.1%
Reasoning, Knowledge, and Epistemics
HLE-Diamond15.2%7.2%9.1%15.3%8.8%
CritPt v2
Full / Hard
31.4/17.0%โ€“/4.0%โ€“/7.0%10.6/4.0%36.9/13.5%
Apex 202534.4%5.2%4.2%18.8%22.9%
ARC-AGI
v1 / v2
84.5/49.3%โ€“โ€“87.5/42.4%22.7/1.5%
LongBench v266.9%52.9%33.1%52.3%62.1%
GraphWalks BFS
256K
69.7%โ€“โ€“50.4%63.3%
MRCR v2
8-needle 256Kโ€“512K
46.4%โ€“โ€“38.0%5.7%
AA-Omniscience v1.1
score
-8.6-48.5-32.2-6.4-2.9
AA-Omniscience v1.1
non-hallucination rate
63.4%14.5%16.4%71.4%69.5%
ComplexConstraints v1.1
openhands v1.45.0
33.4%31.6%35.6%30.1%30.1%
Safety and Alignment
FORTRESS (adversarial)
Public
89.5%67.8%90.3%78.8%79.6%
FORTRESS (benign)
Public
96.4%97.6%89.1%97.8%92.7%
StrongREJECT (none)97.8%98.1%99.0%99.0%99.4%
Misbehavior (Internal) โ†“9.8%27.6%46.4%57.7%52.8%
Covert misbehavior (Internal) โ†“8.3%24.3%45.6%33.2%43.4%
Destructive actions (Internal) โ†“10.2%75.8%68.8%35.2%65.6%

Best reported score in each row (lowest where lower is better); equal displayed scores tieโ€ƒSecond-best reported score in each row; none when first place is tied

6. Safety

We conducted safety evaluations ahead of release, spanning both everyday human-AI interaction and dangerous-capability testing. Because Inkling 2 Mini is multimodal, we paid attention to whether safety behavior held consistently across text and image inputs. We applied mitigations to reduce risks before release, including an extra phase of alignment training using on-policy distillation, similarly to Inkling 2 Small.

For everyday use, we evaluated sycophancy and how the model handles requests involving suicide, self-harm and eating disorders, including whether it encourages harm or affirms false beliefs. We also assessed whether the model refuses genuinely harmful requests without over-refusing benign ones. For CBRN and cyber, we assessed knowledge and procedural uplift through internal evaluations and refusal-suppressed variants intended to estimate latent capability with safeguards removed. For loss of control, we evaluated agentic capability, strategic deception, and sabotage potential, benchmarked against public frontier models, and found the model materially below frontier capabilities.

Inkling 2 Mini is the first looped transformer we have released, so we also studied how looping affects monitorability: whether we can still oversee the model by reading its chain of thought. Looping lets the model do more computation without reasoning in visible tokens. Inkling 2 Mini has very little controllability over its chain of thought, and was unable to perform covert side tasks in realistic agentic environments. Further experiments on the effect of looping on monitorability are described in our blog post.

Across all areas, we concluded that Inkling 2 Mini does not give material uplift beyond what is already available from open-weights models.

7. Bias, risks and limitations

Inkling 2 Mini may exhibit general limitations common to foundation models, including hallucination (generating plausible but factually incorrect or unsupported content), occasional failures to follow instructions precisely, and degraded performance in long multi-turn conversations. As with other models trained on web-derived and synthetic data, it may reflect biases present in its training data, including demographic, cultural or linguistic biases, and may perform unevenly across languages, dialects or subject domains that were less represented during training.

Inkling 2 Miniโ€™s knowledge is limited to information available as of its training cutoff, and it may not reflect events or changes that occurred afterward.

Inkling 2 Mini is a small model built to run on local hardware, and some limitations are specific to it:

  • Knowledge and factual accuracy. It is less capable than Inkling 2 Small across most of our evaluations, with the largest gaps on factual recall and knowledge-heavy questions. It is more likely than larger models to state incorrect information with confidence.
  • Repetitive reasoning. In internal testing the model sometimes became stuck in long reasoning loops, or repeatedly re-checked its own work without making progress, particularly when debugging code. We recommend setting limits on output tokens and agent turns.
  • Agentic reliability. When using tools, it is less reliable than larger models at following detailed instructions and project conventions. It may take destructive actions, such as deleting or overwriting files, without confirmation, and may not report the outcome of its work accurately.
  • Languages other than English. Output quality can be lower outside English. In testing in Chinese we observed clipped phrasing and mixed languages.
  • Sensitive topics. The model can respond with more caution than the situation calls for when a conversation touches on mental health.
  • Reasoning and compute. Setting reasoning effort to its minimum does not fully disable reasoning. Because the middle layers run three times, compute per token is higher than the active-parameter count alone suggests.
  • Local performance. Support for looped models in local runtimes is new, and speed varies by runtime and hardware.

We recommend that downstream developers and deployers:

  • Apply appropriate human oversight and review for outputs used in high-stakes or safety-critical contexts, rather than relying on Inkling 2 Miniโ€™s outputs without verification.
  • Run agentic workloads in a sandbox, and require confirmation before destructive or irreversible actions.
  • Conduct their own evaluation of Inkling 2 Miniโ€™s performance, safety and fairness for their specific use case, language and population prior to deployment, particularly for applications involving vulnerable groups.
  • Implement additional safeguards, such as content filtering, rate limiting and monitoring, at the application layer, especially for open deployment contexts where Inkling 2 Miniโ€™s built-in mitigations may not be sufficient on their own.
  • Avoid deploying Inkling 2 Mini in domains such as medical, legal or safety-critical decision-making without additional fine-tuning, domain-specific validation and human oversight.

8. Legal

Training Data Documentation

Downloads last month
1,388
GGUF
Model size
26B params
Architecture
inkling2_loop
Hardware compatibility
Log In to add your hardware

4-bit

8-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for Xylolabs/Inkling-2-Mini-GGUF

Quantized
(180)
this model

Collection including Xylolabs/Inkling-2-Mini-GGUF