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. Gemini 4 ArgonGoogleRemove
  2. Llama 3.2 1b InstructMetaRemove
  3. Nemotron 3.5 Super VLNVIDIARemove
gemini-4-argon vs llama-3.2-1b-instruct vs nemotron-3.5-super-vl
AttributeGemini 4 Argongemini-4-argonLlama 3.2 1b Instructllama-3.2-1b-instructNemotron 3.5 Super VLnemotron-3.5-super-vl
Pricing
Input$2.00 / 1M$0.0025 / 1M$0.10 / 1M
Output$10.00 / 1M$0.005 / 1M$0.40 / 1M
Cache Write (5m)$2.00 / 1M$0.0025 / 1M$0.10 / 1M
Cache Write (1h)$2.00 / 1M$0.0025 / 1M$0.10 / 1M
Cache Read$2.00 / 1M$0.0025 / 1M$0.10 / 1M
Web Search$0 / 1M$0 / 1M$0 / 1M
Context
Max context1MN/A1M
Max outputN/AN/AN/A
Capabilities
VisionYesNoYes
Function CallingYesYesYes
JSON ModeYesYesYes
StreamingYesYesYes
Catalogue
ProviderGoogleMetaNVIDIA
Categorychatchatchat
Charge typePay As You GoPay As You GoPay As You Go
Released———
Description
SummaryGemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across long, multi-step tasks and to turn long documents mixing text, tables, charts and images into a single grounded output. It leads the Vals Index across finance, legal, coding and tax work, and supports long decode of up to 1M output tokens in a single response.Llama 3.2 1B is a lightweight 1-billion-parameter model built for efficient NLP tasks like summarization, conversation, and multilingual analysis. It runs well in low-resource environments, supports eight core languages (and can be fine-tuned for more), making it a good fit for developers who need capable, multilingual AI without heavy compute costs.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.