Skip to content
Weights Not Walls
Menu

thinkingmachines

Inkling-2-Small

Reviewedaccepts images
Size
594B parametersfrom the GGUF metadata of a conversion, because this repo's own weights are packed: unsloth/Inkling-2-Small-GGUF
Licence
custom, read it, Apache License 2.0Our reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
On the Hub since
4 Nov 2026the date the repo was created
Context length
1M tokensas an API provider reports it
Kind
Base model (no base model declared on the Hub)Inkling2ForConditionalGeneration
On Hugging Face
thinkingmachines/Inkling-2-Small37k downloads in the last 30 days, 560 likes

Scores

5 independent results on 5 eval sets
APEX-AgentsReported

Epoch's name: inkling-2-small. Best closed model on this set: gemini-4-carbon_unknown, 84.2% (Google DeepMind). All 66 results

ARC-AGIReported

Epoch's name: inkling-2-small. Best closed model on this set: claude-fable-5_max, 98.5% (Anthropic). All 318 results

ARC-AGI-2Reported

Epoch's name: inkling-2-small. Best closed model on this set: gpt-6-astra_max, 95.0% (OpenAI). All 299 results

FrontierSWEReported

Epoch's name: inkling-2-small. Best closed model on this set: claude-fable-5-2_max, 69.9% (Anthropic). All 25 results

GDP.pdfReported

Epoch's name: inkling-2-small. Best closed model on this set: gemini-4-carbon_high, 56.4% (Google DeepMind). All 19 results

Claimed by the maker Claimed

The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.

Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.

Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.

Run it yourself

file sizes read from the Hub 19 Nov 2026

Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 1.5 GB, worked out from the model's config (thinkingmachines/Inkling-2-Small), at 16-bit precision. Choose a machine above to see a verdict for each format. Method

GGUF for llama.cpp, Ollama, LM Studio and others

unsloth/Inkling-2-Small-GGUF (converted by unsloth), 6.4k downloads. Sizes are the real files on the Hub, all parts added up. For image input you also need the projector file mmproj-F16.gguf (0.61 GB), not counted below.

FormatSizeMemory neededGet it
UD-IQ1_S159.9 GBfile size, 5 parts161.9 GB159.9 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-IQ1_S
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-IQ1_S/Inkling-2-Small-UD-IQ1_S-*-of-00005.gguf" --local-dir Inkling-2-Small-GGUF
UD-IQ1_M166.3 GBfile size, 5 parts168.3 GB166.3 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-IQ1_M
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-IQ1_M/Inkling-2-Small-UD-IQ1_M-*-of-00005.gguf" --local-dir Inkling-2-Small-GGUF
UD-IQ2_XXS177.5 GBfile size, 5 parts179.5 GB177.5 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-IQ2_XXS
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-IQ2_XXS/Inkling-2-Small-UD-IQ2_XXS-*-of-00005.gguf" --local-dir Inkling-2-Small-GGUF
UD-Q2_K_XL190.9 GBfile size, 6 parts192.9 GB190.9 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-Q2_K_XL
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-Q2_K_XL/Inkling-2-Small-UD-Q2_K_XL-*-of-00006.gguf" --local-dir Inkling-2-Small-GGUF
UD-IQ3_XXS210.5 GBfile size, 6 parts212.5 GB210.5 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-IQ3_XXS
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-IQ3_XXS/Inkling-2-Small-UD-IQ3_XXS-*-of-00006.gguf" --local-dir Inkling-2-Small-GGUF
UD-Q3_K_XL265.5 GBfile size, 7 parts267.5 GB265.5 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-Q3_K_XL
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-Q3_K_XL/Inkling-2-Small-UD-Q3_K_XL-*-of-00007.gguf" --local-dir Inkling-2-Small-GGUF
UD-Q4_K_XL329.8 GBfile size, 9 parts331.8 GB329.8 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-Q4_K_XL
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-Q4_K_XL/Inkling-2-Small-UD-Q4_K_XL-*-of-00009.gguf" --local-dir Inkling-2-Small-GGUF
UD-Q8_K_XL360.0 GBfile size, 9 parts362.0 GB360.0 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:UD-Q8_K_XL
Download
hf download unsloth/Inkling-2-Small-GGUF --include "UD-Q8_K_XL/Inkling-2-Small-UD-Q8_K_XL-*-of-00009.gguf" --local-dir Inkling-2-Small-GGUF
BF16451.1 GBfile size, 11 parts453.1 GB451.1 + 1.5 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Inkling-2-Small-GGUF:BF16
Download
hf download unsloth/Inkling-2-Small-GGUF --include "BF16/Inkling-2-Small-BF16-*-of-00011.gguf" --local-dir Inkling-2-Small-GGUF

MLX for Apple silicon

FormatSizeMemory neededRepo
MLXmlx-community/Inkling-2-Small-oQ4e318.0 GBfile size320.0 GB318.0 + 1.5 + 0.50Download page
MLX 4bitmlx-community/Inkling-2-Small-4bit354.6 GBfile size356.6 GB354.6 + 1.5 + 0.50Download page

Original weights

The maker's own weights are packed into integer tensors, so the safetensors header gives no size we can trust; the repo lists the real file sizes.

hf download thinkingmachines/Inkling-2-Small --local-dir Inkling-2-Small
Other conversions (9; sizes not read)

The Hub tags 21 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.

Run it by API

Hugging Face Inference Providers, fetched 19 Nov 2026, 13:58 UTC

6 providers serve this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.

ProviderInOutContextPrecisionTokens/sFirst tokenUptime 24hToolsJSON
PrimeIntellect$0.15$0.501Mfp4841.4 s99.8%yesyes
DeepInfra$0.28$0.951Mfp41820.6 s99.4%yesyes
Cloudflare$0.30$1.001Mnot given2371.0 s99.9%yesyes
Modal$0.30$1.001Mnvfp41980.8 s100.0%yesyes
Nebius$0.30$1.001Mfp41760.3 s97.3%yesyes
Together$0.30$1.001Mnot given1570.5 s100.0%yesyes

Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.

curl https://router.huggingface.co/v1/chat/completions \ -H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \ -d '{"model": "thinkingmachines/Inkling-2-Small:primeintellect", "messages": [{"role": "user", "content": "Hello"}]}'

Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.

Fine-tunes of this model

from our snapshot of the Hub

Repos that declare this model as their base, most downloaded first. Listing is not a recommendation; unreviewed ones have not been checked.

Planned for this page Not built yet

Try it in private chat
A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
Fine-tune it
Fine-tuning a few base models from this page and publishing the result here.
Creators and earnings
Pages for people who publish fine-tunes and adapters of this model, and what they earn.
Fund and bounties
Bounties for new open models, paid from a public fund. How the fund will work.
Contribute compute
How to give compute to open model work, with Akash.

Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.