selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

thinkingmachines logo

Inkling-2-Mini

thinkingmachines·Instruct·released 17 Nov 2026

Apache 2.026.4BA3.5BMoE1M ctx ≈17 GB
Your GPU
ollama run inkling-2-mini
All 4 runners ↓

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.

How much VRAM does Inkling-2-Mini need?

At Q4_K_M Inkling-2-Mini needs about 17 GB with an 8K context (17 GB at Q4, 53 GB at full precision). Pick your GPU to see what fits and how fast it runs.

--vram
any
--gpus
--quant

Needs: ≈17 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.

Change quant, context or KV cache to compare
Advanced: context 8K · KV cache 16-bit
Context
KV cache

Smallest GPU for each quant and context (KV cache 16-bit)

Quantisation8K16K32K64K128K1M
Q3_K_M
Q4_K_M
Q5_K_M
Q6_K
Q8_0
FP16

≤16 GB card24–32 GB48–96 GBmore

QuantisationQualityWeightsKV cacheTotal
Q2_KHeavy lossSmallest, noticeably worse qualityHeavy loss8.1 GB est.0.8 GB≈9.7 GB
Q3_K_MNoticeable lossSmall, some quality lossNoticeable loss12 GB est.0.8 GB≈14 GB
Q4_K_MrecommendedSmall lossRecommended default for local useSmall loss15 GB est.0.8 GB≈17 GB
Q5_K_MNear-losslessNear-lossless for most tasksNear-lossless17 GB est.0.8 GB≈20 GB
Q6_KNear-losslessPractically losslessNear-lossless20 GB est.0.8 GB≈23 GB
Q8_0LosslessLossless in practice, twice the size of Q4Lossless0.4 GB file0.8 GB≈1.6 GB
FP16 / BF16LosslessFull precision, as releasedLossless0.6 GB file0.8 GB≈1.9 GB

Estimates for 8K tokens of context: weights + KV cache + runtime overhead. All experts stay in memory, but only the active ones are read per token. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it. Weights marked “file” use the real GGUF download size; “est.” rows are computed from the parameter count.

How to pick a quant

Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.

  • LosslessF16, Q8_0
  • Near-losslessQ6_K, Q5_K_M
  • Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
  • Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
  • Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K

Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B

How much context fits?

Model quant
KV cache

At 8K context, Q4_K_M needs ≈17 GB(15 GB weights + 0.8 GB KV cache + 1.2 GB overhead). Pick a GPU above to see how much context fits.

0501001K2K4K8K16K32K64K128K256K512K1MGB

Weights + overheadKV cache

Can the RTX 4060 run Inkling-2-Mini? →

Speed vs quality per quant

Smaller quants run faster and fit in less memory, but lose quality.

Q2_KQ3_K_MQ4_K_MQ5_K_MQ6_KQ8_0F16

Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.

Real-world speed of Inkling-2-Mini

Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →

No medians yet. Be one of the first 3 to report a setup.

All reports

Run Inkling-2-Mini locally

--quantQ4_K_Mdefault; pick your GPU to tailor itunsloth/Inkling-2-Mini-GGUF

From the Ollama librarydocs
ollama run inkling-2-mini

The library build picks its own default quantisation.

UD-Q4_K_M straight from Hugging Facedocs
ollama run hf.co/unsloth/Inkling-2-Mini-GGUF:UD-Q4_K_M

Install Ollama

OpenAI-compatible server with a web UIdocs
llama-server -hf unsloth/Inkling-2-Mini-GGUF:UD-Q4_K_M

Downloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.

Install llama.cpp

Download with the lms CLIdocs
lms get unsloth/Inkling-2-Mini-GGUF@UD-Q4_K_M

Then load it from the LM Studio app or with `lms load`.

Install LM Studio

Ask about an imagedocs
python
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="thinkingmachines/Inkling-2-Mini", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])

Install Transformers

Quantised versions and downloads

RepositoryFormatQuantSizeDownloads
unsloth/Inkling-2-Mini-GGUFGGUFUD-IQ1_S7.8 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-IQ1_M8.0 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-IQ2_XXS8.5 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-Q2_K_XL9.1 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-IQ3_XXS12 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-Q3_K_XL14 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-IQ4_XS16 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-Q4_K_M17 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-Q4_K_XL18 GB—
unsloth/Inkling-2-Mini-GGUFGGUFUD-Q8_K_XL23 GB—

About Inkling-2-Mini

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 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. 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:

Full model card on Hugging Face →

Alternatives to Inkling-2-Mini

All alternatives →
Qwen logo

Qwen3.8-27B

Qwen·Vision

≈19 GB

Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.

27.8B256K ctx Apache 2.0
6.5MAug 2026ollama
google logo

gemma-4-26B-A4B-it

google·Vision

≈18 GB

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.

25.8BA4B256K ctx Apache 2.0
4.9MMar 2026ollama
Qwen logo

Qwen3.6-27B

Qwen·Vision

≈19 GB

Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding…

27.8B256K ctx Apache 2.0
2.4MApr 2026ollama
google logo

gemma-4-31B-it

google·Vision

≈27 GB

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.

31.3B256K ctx Apache 2.0
6.1MMar 2026ollama
google logo≈18 GB

DiffusionGemma is a generative model built by Google DeepMind. Based on the 26B A4B Mixture-of-Experts (MoE) Gemma 4 architecture, DiffusionGemma generates tokens using discrete diffusion. This open-weights model is multimodal, handling text, image, and video inputs to generate text output.

25.8BA4B256K ctx Apache 2.0
581KJun 2026gguf
Qwen logo

Qwen3.5-27B

Qwen·Vision

≈19 GB

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.

27.8B256K ctx Apache 2.0
1.6MFeb 2026ollama

Inkling-2-Mini FAQ

How much VRAM does Inkling-2-Mini need?
About 17 GB at Q4_K_M with an 8K context. Q8_0 needs about 29 GB and full precision about 53 GB. Longer contexts need more memory for the KV cache.
Can Inkling-2-Mini run on the RTX 4060?
Yes. It fits at FP16 / BF16 (about 1.9 GB), at an estimated 22 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.
Can I use Inkling-2-Mini commercially?
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.
What is the context length of Inkling-2-Mini?
1,048,576 tokens (1M). The full window needs far more memory than an 8K context.
How do I run Inkling-2-Mini with Ollama or llama.cpp?
With Ollama: ollama run inkling-2-mini. With llama.cpp: llama-server -hf unsloth/Inkling-2-Mini-GGUF:UD-Q4_K_M, which starts an OpenAI-compatible server.

Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.