Inkling Small (276B-A12B) on NVIDIA H100 NVL

Can the NVIDIA H100 NVL run Inkling Small (276B-A12B)?

Yes, it runs well (offloads to RAM)

Runs well (offloads to RAM). At Q2_K it needs about 112 GB of 94.0 GB usable, with an estimated speed of 51.4 tok/s.

Best quant: Q2_K
Generation speed: 51.4 tok/s
Prompt processing: 6,349 tok/s
First token (1K prompt): 0.3 s
Largest safe context: —

Best app to run it on the NVIDIA H100 NVL

On Linux

LM Studio

Desktop app · runs on llama.cpp, MLX

Splits the model between GPU memory and system RAM, which this setup needs.
Point-and-click: search, download and chat in one window.

Open Inkling Small (276B-A12B) in LM Studio

llama.cpp Splits the model between GPU memory and system RAM, which this setup needs.
Ollama Splits the model between GPU memory and system RAM, which this setup needs.

On Windows

LM Studio

Desktop app · runs on llama.cpp, MLX

Splits the model between GPU memory and system RAM, which this setup needs.
Point-and-click: search, download and chat in one window.

Open Inkling Small (276B-A12B) in LM Studio

llama.cpp Splits the model between GPU memory and system RAM, which this setup needs.
Ollama Splits the model between GPU memory and system RAM, which this setup needs.

Want an agent? Inkling Small (276B-A12B) supports tool use, so you can pair a llama.cpp or MLX LM server with one of these. Apps that run on this computer come first.

Coding agents

Personal assistants

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Inkling Small (276B-A12B)

Recommended for the NVIDIA H100 NVL: Q2_K. Every other quantization has its own download link in the table below.

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q2_K (UD-Q2_K_XL) files (opens in a new tab)

huggingface.co/unsloth/Inkling-Small-GGUF (opens in a new tab)

ollama run hf.co/unsloth/Inkling-Small-GGUF

Original weights · Hugging Face

Full-precision files from Thinking Machines Lab, for fine-tuning, vLLM, Transformers.

huggingface.co/thinkingmachines/Inkling-Small (opens in a new tab)

No sign-in needed.

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 65Good quality (B) · 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q2_K, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs welloffloads to RAM: How well it runs on the NVIDIA H100 NVL: speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	29.5 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	10.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Match score	69.5 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the NVIDIA H100 NVL

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	112 GB	94.0 GB	51.4 tok/s	Runs welloffloads to RAM	65Good quality (B) · 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (UD-Q2_K_XL) · folder (opens in a new tab)
`Q3_K_M`	136 GB	94.0 GB	—	Won't run	78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 10/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q4_K_M`	168 GB	94.0 GB	—	Won't run	89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q5_K_M`	194 GB	94.0 GB	—	Won't run	90Excellent quality (A+) · 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q6_K`	224 GB	94.0 GB	—	Won't run	92Excellent quality (A+) · 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`Q8_0`	287 GB	94.0 GB	—	Won't run	93Excellent quality (A+) · 93 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab)
`F16`	537 GB	94.0 GB	—	Won't run	93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 12/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · folder (opens in a new tab) Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

92Excellent quality (A+) · 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (512 GB) Q6_K · 42.4 tok/s · Runs great
89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (256 GB) Q4_K_M · 58.0 tok/s · Runs well (tight fit)
78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. Apple M2 Ultra (192 GB) Q3_K_M · 70.3 tok/s · Runs well (tight fit)
65Good quality (B) · 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q2_K · 51.5 tok/s · Runs well (offloads to RAM)
65Good quality (B) · 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GB10 (128 GB) Q2_K · 34.1 tok/s · Runs well (over GPU memory limit)
93Excellent quality (A+) · 93 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. Apple M5 Ultra (512 GB) Q8_0 · 44.7 tok/s · Runs great

Quick answers

How much memory does Inkling Small (276B-A12B) need on the NVIDIA H100 NVL 94GB?

At Q2_K it needs about 112 GB. The NVIDIA H100 NVL 94GB has 94.0 GB usable.

How fast is Inkling Small (276B-A12B) on the NVIDIA H100 NVL 94GB?

About 51.4 tok/s at Q2_K, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Inkling Small (276B-A12B) should I use on the NVIDIA H100 NVL 94GB?

Q2_K. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.