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. Qwen3 Coder FlashAlibabaRemove
  3. Nemotron 3.5 Super VLNVIDIARemove
nemotron-3.5-super-vl:free vs qwen3-coder-flash vs nemotron-3.5-super-vl
AttributeNemotron 3.5 Super VL (Free)nemotron-3.5-super-vl:freeQwen3 Coder Flashqwen3-coder-flashNemotron 3.5 Super VLnemotron-3.5-super-vl
Pricing
Input$0 / 1M$0.30 / 1M$0.10 / 1M
Output$0 / 1M$1.50 / 1M$0.40 / 1M
Cache Write$0 / 1M——
Cache Read$0 / 1M$0.30 / 1M$0.10 / 1M
Cache Write (5m)—$0.30 / 1M$0.10 / 1M
Cache Write (1h)—$0.30 / 1M$0.10 / 1M
Web Search—$0 / 1M$0 / 1M
Context
Max context1M128K1M
Max outputN/AN/AN/A
Capabilities
VisionYesNoYes
Function CallingYesYesYes
JSON ModeYesYesYes
StreamingYesYesYes
Catalogue
ProviderNVIDIAAlibabaNVIDIA
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.Qwen3 Coder Flash is Alibaba's fast, cost-efficient coding agent model — a lighter version of Qwen3 Coder Plus — built for autonomous programming through tool calling and environment interaction, while still retaining strong general-purpose abilities.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.