Compare models
Put up to 4 models beside each other — token prices, context windows, capabilities and provider, from the same catalogue the model pages read.
| Attribute | Nemotron 3.5 Super VL (Free)nemotron-3.5-super-vl:free | Perceptron Mk1.5perceptron-mk1.5 | Nemotron 3.5 Lightning (Free)nemotron-3.5-lightning:free |
|---|---|---|---|
| Pricing | |||
| Input | $0 / 1M | $0.15 / 1M | $0 / 1M |
| Output | $0 / 1M | $1.50 / 1M | $0 / 1M |
| Cache Write | $0 / 1M | — | $0 / 1M |
| Cache Read | $0 / 1M | $0.15 / 1M | $0 / 1M |
| Cache Write (5m) | — | $0.15 / 1M | — |
| Cache Write (1h) | — | $0.15 / 1M | — |
| Web Search | — | $0 / 1M | — |
| Context | |||
| Max context | 1M | 36.9K | 1M |
| Max output | N/A | N/A | N/A |
| Capabilities | |||
| Vision | Yes | No | Yes |
| Function Calling | Yes | No | Yes |
| JSON Mode | Yes | No | Yes |
| Streaming | Yes | Yes | Yes |
| Catalogue | |||
| Provider | NVIDIA | Perceptron | NVIDIA |
| Category | chat | chat | chat |
| Charge type | Free | Pay As You Go | Free |
| Released | — | 2026-09-25 | — |
| Description | |||
| Summary | NVIDIA Nemotron 3.5 Super VL is an open multimodal Mixture-of-Experts (MoE) model featuring 120B total parameters with 12B activated per token. Built on a hybrid Mamba–Transformer architecture with multi-token prediction, it supports text, images, documents, GUI screenshots, and video inputs, along with a 1M-token context window for complex, long-horizon workflows. Optimized for multimodal reasoning, visual understanding, and agentic tool use, the model supports configurable reasoning modes and token budgets. It is well suited for document intelligence, web research, coding validation, customer service, GUI analysis, and video understanding, enabling efficient, long-running multimodal agent applications. | Perceptron Mk1.5 chat model that accepts audio input. Audio and text input are billed per token. | NVIDIA Nemotron 3.5 Lightning is an open Mixture-of-Experts (MoE) model with 30B total parameters and 3B active per token, optimized for high-throughput agentic workloads and efficient inference. Its lightweight active compute and open design make it well suited for specialized agents, domain-specific customization, and scalable production deployments where speed, cost efficiency, and adaptability are key. |