Skip to content

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.

  1. Nemotron 3.5 Super VL (Free)NVIDIARemove
  2. Ministral 3 14B 2512Mistral AIRemove
  3. Nemotron 3.5 Super VLNVIDIARemove
nemotron-3.5-super-vl:free vs ministral-14b-2512 vs nemotron-3.5-super-vl
AttributeNemotron 3.5 Super VL (Free)nemotron-3.5-super-vl:freeMinistral 3 14B 2512ministral-14b-2512Nemotron 3.5 Super VLnemotron-3.5-super-vl
Pricing
Input$0 / 1M$0.20 / 1M$0.10 / 1M
Output$0 / 1M$0.20 / 1M$0.40 / 1M
Cache Write$0 / 1M——
Cache Read$0 / 1M$0.20 / 1M$0.10 / 1M
Cache Write (5m)—$0.20 / 1M$0.10 / 1M
Cache Write (1h)—$0.20 / 1M$0.10 / 1M
Web Search—$0 / 1M$0 / 1M
Context
Max context1M262.1K1M
Max outputN/AN/AN/A
Capabilities
VisionYesYesYes
Function CallingYesYesYes
JSON ModeYesYesYes
StreamingYesYesYes
Catalogue
ProviderNVIDIAMistral AINVIDIA
Categorychatchatchat
Charge typeFreePay As You GoPay As You Go
Released———
Description
SummaryNVIDIA 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.Ministral 3 14B is the largest model in the Ministral 3 lineup, delivering near–frontier performance similar to the larger Mistral Small 3.2 24B. It's a powerful yet efficient language model that also includes vision capabilities.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.