nvidia/nemotron-3.5-super-vl

Model release OpenRouter 模型 Source language: Chinese

OpenRouter 于 2026-10-15 收录该模型(收录时间不等于模型正式发布日期);上下文:1,048,576 tokens;输入价:$0.1500/百万 tokens;输出价:$0.5000/百万 tokens;定位(来源描述):NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B…

Full text

For the free endpoint, please do not upload any confidential information or personal data (such as voices or faces of people). Your use is logged for security purposes and to improve NVIDIA products and services. The logged session data for improvement purposes is not linked to your identity or any persistent identifier. For more information about NVIDIA's data processing practices, see Privacy PolicyOpens in new tab. By using this free endpoint, you consent to NVIDIA's collection, recording, and use of such information and the NVIDIA API Trial Terms of ServiceOpens in new tab

NVIDIA: Nemotron 3.5 Super VL

nvidia/nemotron-3.5-super-vl

Model weights

Compare

NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video as input and returns text, and is built for long-running agents that combine visual understanding with reasoning and tool use.

It uses a hybrid Mamba-Transformer architecture with multi-token prediction and supports a 1M-token context. Typical uses include document intelligence, web research, coding validation, customer service, and video understanding. Reasoning is on by default and can be turned off or given a token budget.

Modalities

In / out price

$0.06 / $0.35per 1M

Context

1.0M

Released

Oct 15, 2026

Compare

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput

$0.15$0.50$0.0200.59s155 tok/s

99.78%

$0.15$0.50$0.0201.19s166 tok/s

99.89%

$0.15$0.50$0.0201.26s124 tok/s

99.84%

40% off

$0.15$0.09$0.50$0.30$0.02$0.0123.38s59 tok/s

99.58%

$0.30$0.90$0.0602.03s108 tok/s

99.61%

$0.06$0.35$0.0101.90s23 tok/s

98.25%

$0.10$0.40$0.0806.43s28 tok/s

97.87%

$0.15$0.47$0.0200.67s394 tok/s

96.85%

$0.20$0.50$0.0400.34s373 tok/s

100.00%

$0.20$0.50$0.0401.50s181 tok/s

100.00%

ProviderInput / Output /M

Input $0.15 / Output $0.50

155 tok/s99.78% uptime

Input $0.15 / Output $0.50

166 tok/s99.89% uptime

Input $0.15 / Output $0.50

124 tok/s99.84% uptime

Input $0.09 / Output $0.30

40% off

59 tok/s99.58% uptime

Input $0.30 / Output $0.90

108 tok/s99.61% uptime

Input $0.06 / Output $0.35

23 tok/s98.25% uptime

Input $0.10 / Output $0.40

28 tok/s97.87% uptime

Input $0.15 / Output $0.47

394 tok/s96.85% uptime

Input $0.20 / Output $0.50

373 tok/s100.00% uptime

Input $0.20 / Output $0.50

181 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.05444/M tokens

Output

$0.5579/M tokens

Nemotron 3.5 Super VL — Price history

Price history

No effective price data for this selection

$0.0177$0.29992.7%2.0%

$0.0277$0.47094.0%8.1%

$0.0342$0.50089.0%6.9%

$0.0390$0.50085.4%57.6%

$0.0463$0.50079.7%1.3%

$0.0483$0.34823.3%0.3%

$0.0743$0.50078.5%0.5%

$0.0933$0.40033.6%11.2%

$0.1307$0.90070.5%10.5%

$0.1454$0.50034.1%1.5%

ProviderInput / Output /M

Token share

Effective in $0.0177 / Effective out $0.299

Token share

2.0%

Effective in $0.0277 / Effective out $0.470

Token share

8.1%

Effective in $0.0342 / Effective out $0.500

Token share

6.9%

Effective in $0.0390 / Effective out $0.500

Token share

57.6%

Effective in $0.0463 / Effective out $0.500

Token share

1.3%

Effective in $0.0483 / Effective out $0.348

Token share

0.3%

Effective in $0.0743 / Effective out $0.500

Token share

0.5%

Effective in $0.0933 / Effective out $0.400

Token share

11.2%

Effective in $0.1307 / Effective out $0.900

Token share

10.5%

Effective in $0.1454 / Effective out $0.500

Token share

1.5%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

394tok/s

P50, best across providers

Latency

0.34s

P50, best provider

Latency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-Benchio.net92.3%70.5%DeepInfra87.9%74.9%Nebius Token Factory87.5%74.2%CoreWeave88.4%72.4%Together87.4%72.7%

+6 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

100.00%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

99.69%

Availability over the last 3 days

Last 72 hours

Availability 99.69%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.17%

Without Routing

94.43%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for NVIDIA: Nemotron 3.5 Super VL (Artificial Analysis and Design Arena)

SourceBenchmarkScore

Artificial AnalysisNemotron 3.5 Super VL Intelligence Index13.6

Artificial AnalysisNemotron 3.5 Super VL HLE30.7%

Artificial AnalysisNemotron 3.5 Super VL AA-LCR69.0%

Artificial AnalysisNemotron 3.5 Super VL GDPval-AA33.1%

Artificial AnalysisNemotron 3.5 Super VL CritPt10.6%

Artificial AnalysisNemotron 3.5 Super VL SciCode47.8%

Artificial AnalysisNemotron 3.5 Super VL Terminal-Bench 4.00.0%

Artificial AnalysisNemotron 3.5 Super VL AA-Omniscience Accuracy21.2%

Artificial AnalysisNemotron 3.5 Super VL AA-Omniscience Non-Hallucination Rate69.5%

Design ArenaNemotron 3.5 Super VL Models Arena 3D Elo1118

Design ArenaNemotron 3.5 Super VL Models Arena Code Categories Elo1140

Design ArenaNemotron 3.5 Super VL Models Arena Data Visualization Elo1083

Design ArenaNemotron 3.5 Super VL Models Arena Game Development Elo1081

Design ArenaNemotron 3.5 Super VL Models Arena UI Component Elo1155

Design ArenaNemotron 3.5 Super VL Models Arena Website Elo1160

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

27Btokens

2.

Namazu Data Factory

new

19.5Btokens

3.

OpenHands

AI coding agent that can run commands, browse the web, call APIs

10.1Btokens

4.

cyber-train nexus eval

new

8.1Btokens

5.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

6.17Btokens

Rest of OpenRouter

71B

Top 5 Apps

799M

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Not currently available through EU in-region routing. In-region routing docs

Frequently asked questions

NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video as input and returns text, and is built for long-running agents that combine visual understanding with reasoning and tool use.

Nemotron 3.5 Super VL costs $0.06/M input tokens and $0.35/M output tokens, with separate rates for Cache Read at $0.01/M tokens.

Nemotron 3.5 Super VL has a 1,048,576 token context window. It supports up to 131,072 completion tokens.

Yes. Nemotron 3.5 Super VL accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Nemotron 3.5 Super VL accepts text, images, and video as input and returns text.

Nemotron 3.5 Super VL is served by 10 providers on OpenRouter: Darkbloom, io.net, Crusoe, CoreWeave, Prime Intellect, Together, DeepInfra, Venice and 2 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

Nemotron 3.5 Super VL was released on October 15, 2026.

Sources

← Back to the daily brief