Inkling-Small
Inkling-Small is an open-weights language model from Thinking Machines, released July 15, 2026. It has 266B parameters (mixture-of-experts, 2.7B active). On the Epoch Capabilities Index it scores 150.8, rank 67 of 256 rated models.
Today, we are releasing Inkling-Small, an efficient open-weights model that achieves comparable performance to Inkling at a quarter of its size. Inkling-Small is a Mixture-of-Experts transformer with 276B total parameters, 12B active, trained on NVIDIA GB300 NVL72 systems.
Check it against your own PC →
Can you run Inkling-Small locally?
Even with 32 GB of VRAM and 32 GB of RAM it is too large to run at a usable speed on a single consumer GPU. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Too large | – | – | – | – |
| 12 GB VRAM | Too large | – | – | – | – |
| 16 GB VRAM | Too large | – | – | – | – |
| 24 GB VRAM | Too large | – | – | – | – |
| 32 GB VRAM | Too large | – | – | – | – |
Ready-made quantizations on Hugging Face: unsloth/Inkling-Small-GGUF · xautonomics/Inkling-Small-UD-Q4_K_M.LAYR.GGUF · lyssquant/Inkling-Small-GGUF · DevQuasar/thinkingmachines.Inkling-Small-GGUF · miweru/Inkling-Small-REAP-137B-A12B-de-GGUF · Baekpica/Inkling-Small-GGUF
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Math | 36.4 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Knowledge | 65.9 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 90.0% |
| GPQA Diamond | 88.2% |
| FrontierMath (Tiers 1–3) | 48.8% |
| SimpleQA Verified | 19.1% |
| FrontierMath-Tier-4-v2-Private | 17.1% |
| ProofBench | 6.0% |
| Mystery Game Puzzles | 0.0% |
Links
Announcement · Hugging Face: thinkingmachines/Inkling-Small · Epoch AI benchmarks
Models with a similar score
Mistral Large 4 · Qwen3.7-Max · GPT-5.2 · DeepSeek V4 Flash 0731 · GLM-5.3-Flash · Muse Spark · GLM-5.2 · Gemini 3 Pro