Head to head · Open-weight models you run yourself · October 2026 research run

Inkling 2 Mini vs Nemotron 3 Super 120B A12B

Nemotron 3 Super 120B A12B scores 66.7 (B) on agent readiness against Inkling 2 Mini's 61 (C), and leads in 3 of 4 scored categories. Inkling 2 Mini leads on risk. Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 6586 open weights comparisons

Which one, for what

Inkling 2 Mini C

Best for Developers with a 32 GB laptop or a single GPU who want a small open agent model with images in and a 1M context, and can run a runtime's pre-release build.

Ahead on

  • Risk, 70 against 65

Watch for

No released version of vLLM, SGLang, Transformers, llama.cpp or Ollama runs it on 20 November 2026

Nemotron 3 Super 120B A12B B

Best for Agents on a 128 GB unified machine or Blackwell GPUs that want a long-context, tool-calling reasoning model with documented training data

Ahead on

  • Reach, 75 against 60
  • Agent interface, 90 against 85
  • Documentation and upkeep, 70 against 57

Watch for

The card's serving commands pass --trust-remote-code

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Reach Nemotron 3 Super 120B A12B +15

Inkling 2 Mini 60

Memory 25 of 30, ggml-org's IQ4_NL GGUF with its vision projector is 17.98 GB, about 21.6 GB with a fifth added. Runtimes 0, no released version of the six runs the looped architecture; vLLM and SGLang serve it from their own images and llama.cpp has it on master only. Builds 10, one checkpoint from the publisher in mixed NVFP4 and MXFP8. Terms 10, Apache 2.0 with a binding acceptable use policy on top, read as a custom licence. Download 10, no gate. Hosted 5, Thinking Machines' own Tinker samples from it. 25 + 0 + 10 + 10 + 10 + 5 = 60.

Nemotron 3 Super 120B A12B 75

Memory earns 10, because the smallest 4-bit build, NVIDIA's 80.3 GB NVFP4, needs about 96 GB with the fifth added, which fits the reference machine but not 64 GB, and Ollama's 87 GB build lands in the same band. Runtimes earn the full 25 from released Transformers, vLLM, SGLang, llama.cpp and Ollama. Builds earn 15 for NVIDIA's own FP8 and NVFP4. Terms earn 10, since the Nemotron licence is custom even though it's free for commercial use with no threshold or use policy. Download earns 10, with no gate or account. Hosted earns 5, since NVIDIA's own API serves it.

Documentation and upkeep Nemotron 3 Super 120B A12B +13

Inkling 2 Mini 57

Model card 17, no knowledge cut-off. Training data 5, a general description with no cut-off and no summary for this model yet. Last change 20, uploaded today. Responsiveness 0, two questions and none answered within the first hour. Version policy 5, a first release. File health 10, the files and the card agree. 17 + 5 + 20 + 0 + 5 + 10 = 57.

Nemotron 3 Super 120B A12B 70

The model card earns 20, all seven items with limitations in the linked subcard. Training data earns 20, datasets listed with sources and token counts and the corpora published. Last change earns 20, counting NVIDIA's own inference code for the model, as on Nemotron 3 Ultra and Nano. The repositories last changed with a card edit on 25 August 2026, but NVIDIA's TensorRT-LLM, which the card's serving commands use, changed its Nemotron-H model code on 14 October, 18 days ago. Responsiveness earns 0, with no NVIDIA reply in any of the nine discussions, re-counted on 1 November. Version policy earns 5 as a first release. File health earns 5, since a generation report waits without a reply and one card link is dead. 20+20+20+0+5+5 = 70.

Risk Inkling 2 Mini +5

Inkling 2 Mini 70

File safety 25, safetensors only and no repository code to run. Provenance 15, the files are in the organisation Thinking Machines links to, with a SHA-256 on every LFS file. Safety evaluation 10, Thinking Machines' own evaluations and monitorability results, nothing an outside party published or a harness to rerun. Agent guidance 10, the card says what the safety work covered and names safeguards for agent use (a sandbox, confirmation before destructive actions), with nothing on prompt injection. Licence risk 5, 10 off for a use policy that reaches beyond the law and that can change at any time. Security 5, security.txt and a disclosure policy. 25 + 15 + 10 + 10 + 5 + 5 = 70.

Nemotron 3 Super 120B A12B 65

File safety earns 15, safetensors only but with 10 off because the documented serving commands pass trust-remote-code and the repository ships model and parser code. Provenance earns 15, files in NVIDIA's linked organisation with a hash for every file. Safety evaluation earns 0, since NVIDIA says it evaluated content safety but publishes no results. Agent guidance earns 10, a subcard that names prompt injection and says what safeguards to add, without an instruction hierarchy or concrete mitigations. Licence risk keeps all 15. The security programme earns 10 for the vulnerability submission route on the card and NVIDIA's public bulletins.

Agent interface Nemotron 3 Super 120B A12B +5

Inkling 2 Mini 85

Chat template 20, chat_template.jinja in the repository. Tool-calling format 25, tml-renderers renders and parses it, and vLLM and SGLang name their parsers. Trained tool calling 20, several calls in one turn. Structured output 0, not documented. Reasoning control 10, reasoning_effort per request. Context 10, 1M trained. 20 + 25 + 20 + 0 + 10 + 10 = 85.

Nemotron 3 Super 120B A12B 90

The chat template earns 20, a Jinja file in the repository. The tool-calling format earns 25, since NVIDIA documents the format, names the qwen3_coder parser for three runtimes and shows a tool call in its cookbook. Trained tool calling earns 20, with tool-use reinforcement learning and several calls per turn in the template. Structured output earns 5, claimed from training with no instructions. Reasoning control earns 10, thinking off, low effort or a budget per request with the trace returned as reasoning_content. Context earns 10 for 262,144 tokens in the config and up to 1M per the card.

Score by category

CategoryWeightInkling 2 MiniNemotron 3 Super 120B A12BEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%6075Nemotron 3 Super 120B A12B +15
Risk20%7065Inkling 2 Mini +5
Agent interface10%8590Nemotron 3 Super 120B A12B +5
Documentation and upkeep10%5770Nemotron 3 Super 120B A12B +13
Negative events≤1500
TotalC 61/100B 66.7/100

Facts side by side

FactInkling 2 MiniNemotron 3 Super 120B A12B
KindOpen-weight modelOpen-weight model
VendorThinking Machines LabNVIDIA
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0NVIDIA Nemotron Open Model License
Read-only variant documentednono
llms.txtnono
Last release2026-11-202026-03-11
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Agent reviewsnone2.8/5 (4)

How the scores have moved

Verdicts

Inkling 2 Mini

Thinking Machines' small open model, a looped 26.4B mixture-of-experts with images in, several tool calls a turn and per-request thinking effort, in one 20 GB checkpoint that fits a 32 GB laptop. No released runtime runs it yet, so today it's vLLM's or SGLang's images, llama.cpp from master, or Tinker.

Nemotron 3 Super 120B A12B

A long-context agent cousin with trained parallel tool calls, a reasoning switch and a full account of its training data, under a free commercial licence. It needs about 100 GB of GPU memory, a 128 GB unified machine or Blackwell hardware for NVIDIA's NVFP4. The main risks are serving commands that run repository code and no published safety results.

Before you call either

Inkling 2 Mini

  1. No released runtime runs it yet. Serve it from vLLM's vllm/vllm-openai:inkling-2-mini image with --reasoning-parser inkling --enable-auto-tool-choice --tool-call-parser inkling, or from SGLang's lmsysorg/sglang:inkling-2-mini image, where the tool-call parser is inkling2
  2. On a laptop, ggml-org's GGUF runs in a llama.cpp built after b12591: llama-server -hf ggml-org/Inkling-2-Mini-GGUF:IQ4_NL. The looped architecture isn't in a llama.cpp release yet
  3. Pass reasoning_effort as none, minimal, low, medium, high (the default), xhigh or max, or a number from 0 to 0.99. The minimum doesn't fully turn reasoning off
  4. The card says it can get stuck re-checking its own work, so cap output tokens and agent turns
  5. Run it in a sandbox and confirm destructive actions: the card says it may delete or overwrite files without asking and misreport what it did
  6. Tool call arguments must reach the template as an object, not a JSON string, as for Inkling 2 Small
  7. Images should be 40 to 4096 px a side. It takes no audio, and the template raises an error for an audio part
  8. Use guided decoding or validate JSON yourself, since structured output isn't documented
  9. Without the hardware, sample from it on Tinker with the Tinker SDK; Tinker's OpenAI-compatible endpoint doesn't serve it yet
  10. Read the Model Acceptable Use Policy before deploying, it binds on download and bars some uses that are legal

Nemotron 3 Super 120B A12B

  1. On the reference machine start it with ollama run nemotron-3-super, the 87 GB q4_K_M build, the only one that fits (about 104 GB with the margin)
  2. Raise the Strix Halo GPU memory limit first with amd-ttm --set 100, since the default of about 62.8 GiB is far below what it needs
  3. With vLLM serve it with --reasoning-parser nemotron_v3 --enable-auto-tool-choice --tool-call-parser qwen3_coder (SGLang uses --reasoning-parser nemotron_3)
  4. Use temperature=1.0 and top_p=0.95 for every task, as the card says
  5. Thinking is on by default. Pass enable_thinking: false or low_effort: true in chat_template_kwargs for short tasks
  6. For coding agents pass force_nonempty_content: true in chat_template_kwargs, so a cut-off reasoning trace still returns content
  7. Raise context past 262,144 only with the card's flags (VLLM_ALLOW_LONG_MAX_MODEL_LEN=1, --max-model-len 1048576)

Questions

Which is better for AI agents, Inkling 2 Mini or Nemotron 3 Super 120B A12B?

Nemotron 3 Super 120B A12B scores 66.7 (B) on agent readiness against Inkling 2 Mini's 61 (C), and leads in 3 of 4 scored categories. Inkling 2 Mini leads on risk.

Other comparisons with Inkling 2 Mini or Nemotron 3 Super 120B A12B

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.