Spiral gives your agent every tool with one account. Browse the tools →

Codex · LLMReleased 2026-10-15

How to Use Nemotron 3.5 Super VL in Codex

Nemotron 3.5 Super VL reads documents, charts, and short videos and helps investigate coding errors. It combines visual understanding with reasoning and tool use, returning text.

Prompts to try with Nemotron 3.5 Super VL in Codex

Copy an example into Codex, or ask in your own words.

Use Nemotron 3.5 Super VL to inspect this project's failing test and relevant source files. Explain the likely cause, make the smallest justified fix, and run the affected test.

Nemotron 3.5 Super VLCodex
Try asking
  • Use Nemotron 3.5 Super VL to review the current diff. Identify changed behaviors and propose regression tests with explicit expected results.

  • Use Nemotron 3.5 Super VL to compare the project screenshot with the reported UI bug. List visible evidence before proposing code changes; use temperature 0.6.

  • Use Nemotron 3.5 Super VL to extract the table in this project's document image. Return JSON with column names, units, and null for unreadable cells.

  • Use Nemotron 3.5 Super VL to review the project's 60-second MP4 demonstration. Describe the visible failure sequence and identify the code paths worth investigating.

How to set up Nemotron 3.5 Super VL in Codex

Pick the way that suits you. Each takes about two minutes.

Use Spiral's MCP server

  1. Add the hosted server
    codex mcp add spiral --url https://api.spiral.ai/mcp
  2. Sign in

    Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.

    codex mcp login spiral

Use the Spiral skill

  1. Send Codex this message

    Your agent reads Spiral's skill and sets itself up.

    set up https://spiral.ai/SKILL.md
  2. Sign in or paste a key

    Approve Spiral in the browser, or give the agent a key from the console.

Use the Spiral CLI

  1. Install and sign in

    Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

    npx -y spiral-cli@latest init

Call the Spiral API

  1. Create an API key

    In the console, under API keys.

  2. Run Nemotron 3.5 Super VL

    One request; the result comes back by polling or a webhook.

    curl https://api.spiral.ai/v1/runs \
      -H "Authorization: Bearer $SPIRAL_API_KEY" \
      -H "Content-Type: application/json" \
      -H "Idempotency-Key: $(uuidgen)" \
      -d '{
        "provider": "openrouter",
        "model": "nvidia/nemotron-3.5-super-vl",
        "input": {
          "messages": [
            {
              "role": "system"
            }
          ]
        }
      }'

Then paste one of the prompts above: Codex finds Nemotron 3.5 Super VL on Spiral, shows the price and runs it.

Nemotron 3.5 Super VL providers and prices

Spiral runs Nemotron 3.5 Super VL through OpenRouter. Open a card for today's price, with every fee included.

OpenRouterLLM
NVIDIA: Nemotron 3.5 Super VL
nvidia/nemotron-3.5-super-vl

Settings you can ask for

Ask Codex for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Seeda whole number
Min p0 to 1
Top ka whole number from 0
Top p0 to 1
Logprobson or off
Max tokens1 to 943718

The most output tokens to generate.

Temperature0 to 2
Top logprobs0 to 20
Presence penalty-2 to 2
Frequency penalty-2 to 2
Repetition penalty0 to 2

Every API field of NVIDIA: Nemotron 3.5 Super VL is on its tool page: open its card above.

What can you do with Nemotron 3.5 Super VL in Codex?

Turn code, failure logs, and screenshots into a focused debugging investigation with Nemotron 3.5 Super VL. Ask for a proposed fix and a concrete way to check the result.

Investigate a failing test

Ask Nemotron 3.5 Super VL to connect a failure log with the relevant code, propose a fix, and explain how to verify it.

Plan regression coverage

Give Nemotron 3.5 Super VL a code change and ask which behaviors need tests, including edge cases and expected results.

Extract document tables

Ask Nemotron 3.5 Super VL to transcribe a table from a page image, preserve units, and mark unreadable cells.

Explain a dashboard change

Have Nemotron 3.5 Super VL inspect a dashboard screenshot and separate visible changes from possible explanations.

Review a short demonstration

Use Nemotron 3.5 Super VL to summarize the visible sequence in a one-minute product demonstration and list unclear steps.

Compare multilingual explanations

Ask Nemotron 3.5 Super VL to explain the same supplied passage in English and Spanish while preserving names, quantities, and terminology.

Questions about Nemotron 3.5 Super VL in Codex

Can Codex use Nemotron 3.5 Super VL?

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Nemotron 3.5 Super VL like any built-in tool.

How much does Nemotron 3.5 Super VL cost in Codex?

Up to $0.002671 per run at the default inputs. $0.195 per 1M input tokens, $0.65 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

Do I need an OpenRouter account to use Nemotron 3.5 Super VL in Codex?

No. Spiral runs Nemotron 3.5 Super VL on its own OpenRouter account. You sign in to Spiral once and never paste a key.

How long does it take to set up Nemotron 3.5 Super VL in Codex?

About two minutes: add the hosted server, then sign in.

Does Codex wait while Nemotron 3.5 Super VL runs?

Nemotron 3.5 Super VL runs in the background. Codex starts the run, checks on it and gives you the text, json when it's ready.

What is Nemotron 3.5 Super VL useful for?

Use it for questions that combine written evidence with documents, charts, or other visual material. It also supports coding investigations and tool-assisted reasoning.

Does Nemotron 3.5 Super VL generate images or videos?

Its generated output is text. It analyzes visual material and can return descriptions, explanations, or extracted information.

Does 1 million tokens mean a 1-million-token answer?

No. The context window includes the prompt, visual material, conversation, tool results, and generated response; reasoning also uses the output budget.

How long should video clips be?

NVIDIA recommends one- to two-minute MP4 clips. Its documented NIM configuration samples two frames per second and truncates longer clips at 256 frames, so split longer footage into short sections.

Which languages does Nemotron 3.5 Super VL support?

NVIDIA lists English, French, German, Italian, Japanese, Spanish, and Chinese.

Get started

Start using Nemotron 3.5 Super VL in Codex

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Codex:

Set up Codex

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options