Signal SentrySearchUpdated Oct 28, 05:30 UTCAM Drop
Models · scorecard

Nemotron 3.5 Super VL

What the vendor claims, next to what independent boards measured. The two are never mixed in one column.

At a glance

Maker
NVIDIA
Price per 1M tokens
$0.15 in · $0.5 out
Context window
1.05M
Input → output
text, image, video → text
Added to OpenRouter
Oct 15, 2026

Source: OpenRouter, as of Oct 28, 09:19 UTC. Model page on OpenRouter

What the vendor claims

Quoted word for word from the vendor's own launch post or model card. Claims, not measurements.

Scores the vendor claims, each with its source and the quoted line
BenchmarkClaimedSource
CharXivThe source says: On CharXiv, a chart-reasoning benchmark, the model scored 80.8%, compared with 63% for Nemotron 3 Nano Omni.80.8%NVIDIA Developer · Oct 15From Drop #026 · Thu, Oct 15 · Evening
CharXivSetting: with Python code interpretationThe source says: With Python code interpretation, it reached 84%.84%NVIDIA Developer · Oct 15From Drop #026 · Thu, Oct 15 · Evening

Independent results

Measured by independent sources (Epoch AI, LMArena, OpenRouter's benchmark runs, τ²-bench, Vectara), not by the vendor. "Listed as" is the source's own name for the model, so you can check our match.

Independent benchmark results for this model, each with its source and date
BoardResultRank
OpenRouter · GPQA DiamondListed as: Nemotron 3.5 Super VLvia Io Net, the most accurate providerSource: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source93.5%Accuracy12
OpenRouter · GPQA Diamond: cost per taskListed as: Nemotron 3.5 Super VLvia Io Net, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source$0.015Cost per task15
OpenRouter · GPQA Diamond: time per taskListed as: Nemotron 3.5 Super VLvia Io Net, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source395 sTime per task54
OpenRouter · τ²-bench airlineListed as: Nemotron 3.5 Super VLvia Darkbloom, the most accurate providerSource: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source75.5%Accuracy43
OpenRouter · τ²-bench airline: cost per taskListed as: Nemotron 3.5 Super VLvia Darkbloom, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source$0.015Cost per task7
OpenRouter · τ²-bench airline: time per taskListed as: Nemotron 3.5 Super VLvia Darkbloom, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Oct 28, 2026. Source14.6 minTime per task58

Price history

OpenRouter's list price per 1M tokens in each of our snapshots.

  • Output now $0.5
  • Input now $0.15

Changed 2 times across our 26 snapshots.

Show the numbers
This model's OpenRouter price in each of our snapshots
SnapshotInput $/1MOutput $/1M
Oct 28, 09:22 UTC$0.15$0.5
Oct 27, 16:03 UTC$0.15$0.5
Oct 27, 04:03 UTC$0.15$0.5
Oct 26, 16:04 UTC$0.15$0.5
Oct 26, 04:03 UTC$0.15$0.5
Oct 25, 16:04 UTC$0.15$0.5
Oct 25, 06:03 UTC$0.15$0.5
Oct 24, 15:04 UTC$0.15$0.5
Oct 24, 03:03 UTC$0.15$0.5
Oct 23, 15:03 UTC$0.15$0.5
Oct 23, 03:04 UTC$0.15$0.5
Oct 22, 15:03 UTC$0.15$0.5
Oct 22, 05:04 UTC$0.15$0.5
Oct 21, 15:03 UTC$0.15$0.5
Oct 21, 03:02 UTC$0.15$0.5
Oct 20, 15:04 UTC$0.15$0.5
Oct 20, 03:03 UTC$0.15$0.5
Oct 19, 15:04 UTC$0.1$0.4
Oct 19, 10:50 UTC$0.1$0.4
Oct 18, 15:03 UTC$0.1$0.4
Oct 18, 03:05 UTC$0.1$0.4
Oct 17, 15:03 UTC$0.1$0.4
Oct 17, 03:03 UTC$0.1$0.4
Oct 16, 15:02 UTC$0.15$0.5
Oct 16, 03:03 UTC$0.15$0.5
Oct 15, 17:02 UTC$0.15$0.5

Source: OpenRouter, as of Oct 28, 09:22 UTC

In the Drops

Models

Nvidia releases Nemotron 3.5 Super VL and Bolt open models for agents

TechCrunch reports Super VL is a 120-billion-parameter open-weight model for enterprise agents, plus Bolt models for local PCs. Super VL and Bolt 74B-A9B are available now; Bolt E4B later this month.

Why it matters: Nvidia names Skydance, EY, Ramp, CrowdStrike, Perplexity and Salesforce as Super VL users. Teams can post-train it on their own data and run it on their own hardware.

TechCrunch · Oct 19OpenRouter · Oct 15
From Drop #035 · Tue, Oct 20 · Morning
Models

Nvidia releases Nemotron 3.5 Super VL, an open 120B multimodal model for agents

Nvidia says the open model has 120B parameters with 12B active, reads images, documents and video, and gets up to 5x the throughput of open peers on one GPU. OpenRouter lists it at $0.15 in / $0.5 out per 1M tokens.

Why it matters: Open weights let teams post-train it on their own data and run it on their own hardware; Nvidia says Scale AI, Ramp and AI21 have. Bolt 74B-A9B for local agents also shipped.

NVIDIA Blog · Oct 15NVIDIA Developer · Oct 15OpenRouter · Oct 15
From Drop #026 · Thu, Oct 15 · Evening