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
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.
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).
| $0.15 | $0.47 | $0.020 | 0.68s | 214 tok/s | ||
| $0.15 | $0.50 | $0.020 | 0.83s | 147 tok/s | ||
| $0.20 | $0.50 | $0.040 | 0.80s | 210 tok/s | ||
| $0.15 | $0.50 | $0.020 | 1.63s | 104 tok/s | ||
| $0.15 | $0.50 | $0.020 | 0.70s | 77 tok/s | ||
| $0.30 | $0.90 | $0.060 | 2.19s | 93 tok/s | ||
| $0.06 | $0.35 | $0.010 | 3.50s | 31 tok/s | ||
| $0.25 | $1.50 | $0.050 | 0.70s | 77 tok/s | ||
| $0.20 | $0.50 | $0.040 | 1.04s | 338 tok/s | ||
| $0.75 | $2.30 | $0.165 | 9.03s | 9 tok/s | ||
| $0.10 | $0.40 | $0.080 | 3.71s | 37 tok/s |
- Input /M
- $0.15
- Output /M
- $0.47
- Cache read /M
- $0.02
- Latency
- 0.68s
- Throughput
- 214 tok/s
- Uptime
- 99.37%
- Input /M
- $0.15
- Output /M
- $0.50
- Cache read /M
- $0.02
- Latency
- 0.83s
- Throughput
- 147 tok/s
- Uptime
- 100.00%
- Input /M
- $0.20
- Output /M
- $0.50
- Cache read /M
- $0.04
- Latency
- 0.80s
- Throughput
- 210 tok/s
- Uptime
- 99.65%
- Input /M
- $0.15
- Output /M
- $0.50
- Cache read /M
- $0.02
- Latency
- 1.63s
- Throughput
- 104 tok/s
- Uptime
- 98.35%

- Input /M
- $0.15
- Output /M
- $0.50
- Cache read /M
- $0.02
- Latency
- 0.70s
- Throughput
- 77 tok/s
- Uptime
- 99.79%
- Input /M
- $0.30
- Output /M
- $0.90
- Cache read /M
- $0.06
- Latency
- 2.19s
- Throughput
- 93 tok/s
- Uptime
- 99.92%

- Input /M
- $0.06
- Output /M
- $0.35
- Cache read /M
- $0.01
- Latency
- 3.50s
- Throughput
- 31 tok/s
- Uptime
- 44.12%

- Input /M
- $0.25
- Output /M
- $1.50
- Cache read /M
- $0.05
- Latency
- 0.70s
- Throughput
- 77 tok/s
- Uptime
- 83.72%
- Input /M
- $0.20
- Output /M
- $0.50
- Cache read /M
- $0.04
- Latency
- 1.04s
- Throughput
- 338 tok/s
- Uptime
- 99.96%
- Input /M
- $0.75
- Output /M
- $2.30
- Cache read /M
- $0.165
- Latency
- 9.03s
- Throughput
- 9 tok/s
- Uptime
- 89.86%
- Input /M
- $0.10
- Output /M
- $0.40
- Cache read /M
- $0.08
- Latency
- 3.71s
- Throughput
- 37 tok/s
- Uptime
- 98.75%
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
Nemotron 3.5 Super VL — Price history
| $0.0374 | $0.500 | 86.6% | 20.5% | ||
| $0.0409 | $0.470 | 83.9% | 31.7% | ||
| $0.0423 | $0.500 | 82.9% | 3.8% | ||
| $0.0435 | $0.350 | 32.8% | 0.0% | ||
| $0.0535 | $0.500 | 74.3% | 13.3% | ||
| $0.0676 | $0.500 | 82.8% | 1.6% | ||
| $0.0715 | $0.500 | 80.3% | 1.9% | ||
| $0.0878 | $0.400 | 61.1% | 10.3% | ||
| $0.1260 | $0.900 | 72.5% | 11.2% | ||
| $0.5103 | $1.987 | 29.0% | 5.7% | ||
| $0.0781 | $1.500 | 85.9% | 0.1% |
- Effective in $0.0374 / Effective out $0.500Token share20.5%
- Effective in $0.0409 / Effective out $0.470Token share31.7%
- Effective in $0.0423 / Effective out $0.500Token share3.8%
- Effective in $0.0435 / Effective out $0.350Token share0.0%
- Effective in $0.0535 / Effective out $0.500Token share13.3%
- Effective in $0.0676 / Effective out $0.500Token share1.6%
- Effective in $0.0715 / Effective out $0.500Token share1.9%
- Effective in $0.0878 / Effective out $0.400Token share10.3%
- Effective in $0.1260 / Effective out $0.900Token share11.2%
- Effective in $0.5103 / Effective out $1.987Token share5.7%
- Effective in $0.0781 / Effective out $1.500Token share0.1%
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).
P50, best across providers
P50, best provider
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.
Availability over the last 3 days
Last 72 hoursAvailability over the last 24 hours
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.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | Nemotron 3.5 Super VL Intelligence Index | 13.6 |
| Artificial Analysis | Nemotron 3.5 Super VL Finance & Accounting Index | 26.7 |
| Artificial Analysis | Nemotron 3.5 Super VL Strategy & Ops Index | 21.8 |
| Artificial Analysis | Nemotron 3.5 Super VL Legal Index | 25.3 |
| Artificial Analysis | Nemotron 3.5 Super VL Healthcare & Medical Index | 21.0 |
| Artificial Analysis | Nemotron 3.5 Super VL Engineering Index | 23.5 |
| Artificial Analysis | Nemotron 3.5 Super VL Economics Index | 26.1 |
| Artificial Analysis | Nemotron 3.5 Super VL Openness Index | 83.3 |
| Artificial Analysis | Nemotron 3.5 Super VL AutomationBench-AA | 14.8% |
| Artificial Analysis | Nemotron 3.5 Super VL GDP.pdf | 8.0% |
| Artificial Analysis | Nemotron 3.5 Super VL HLE | 30.7% |
| Artificial Analysis | Nemotron 3.5 Super VL AA-LCR | 69.0% |
| Artificial Analysis | Nemotron 3.5 Super VL GDPval-AA | 33.1% |
| Artificial Analysis | Nemotron 3.5 Super VL CritPt | 10.6% |
| Artificial Analysis | Nemotron 3.5 Super VL HLE-Diamond | 8.8% |
| Artificial Analysis | Nemotron 3.5 Super VL SciCode | 47.8% |
| Artificial Analysis | Nemotron 3.5 Super VL Terminal-Bench 4.0 | 0.0% |
| Artificial Analysis | Nemotron 3.5 Super VL FrontierCode-AA 2.0 | 6.1% |
| Artificial Analysis | Nemotron 3.5 Super VL FrontierCode 1.1 | 25.5% |
| Artificial Analysis | Nemotron 3.5 Super VL Terminal-Bench-Science | 0.0% |
| Artificial Analysis | Nemotron 3.5 Super VL CritPt v2 (Hard Subset) | 13.5% |
| Artificial Analysis | Nemotron 3.5 Super VL CritPt v2 | 36.9% |
| Artificial Analysis | Nemotron 3.5 Super VL AA-Omniscience Accuracy | 21.2% |
| Artificial Analysis | Nemotron 3.5 Super VL AA-Omniscience Non-Hallucination Rate | 69.5% |
| Artificial Analysis | Nemotron 3.5 Super VL MLCR-AA | 13.3% |
| Design Arena | Nemotron 3.5 Super VL Agents Arena Pdf Elo | 973 |
| Design Arena | Nemotron 3.5 Super VL Models Arena 3D Elo | 1108 |
| Design Arena | Nemotron 3.5 Super VL Models Arena Code Categories Elo | 1131 |
| Design Arena | Nemotron 3.5 Super VL Models Arena Data Visualization Elo | 1081 |
| Design Arena | Nemotron 3.5 Super VL Models Arena Game Development Elo | 1077 |
| Design Arena | Nemotron 3.5 Super VL Models Arena UI Component Elo | 1154 |
| Design Arena | Nemotron 3.5 Super VL Models Arena Website Elo | 1155 |
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.
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 11 providers on OpenRouter: Darkbloom, Crusoe, CoreWeave, Prime Intellect, Together, DeepInfra, Venice, Baseten and 3 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.