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 CarbonGoogleRemove
  2. Nemotron 3.5 Super VL (Free)NVIDIARemove
  3. Codestral 2508Mistral AIRemove
gemini-4-carbon vs nemotron-3.5-super-vl:free vs codestral-2508
AttributeGemini 4 Carbongemini-4-carbonNemotron 3.5 Super VL (Free)nemotron-3.5-super-vl:freeCodestral 2508codestral-2508
Pricing
Input$2.00 / 1M$0 / 1M$0.45 / 1M
Output$10.00 / 1M$0 / 1M$1.35 / 1M
Cache Write (5m)$2.00 / 1M—$0.45 / 1M
Cache Write (1h)$2.00 / 1M—$0.45 / 1M
Cache Read$2.00 / 1M$0 / 1M$0.45 / 1M
Web Search$0 / 1M—$0 / 1M
Cache Write—$0 / 1M—
Context
Max context1M1M256K
Max outputN/AN/AN/A
Capabilities
VisionYesYesNo
Function CallingYesYesYes
JSON ModeYesYesYes
StreamingYesYesYes
Catalogue
ProviderGoogleNVIDIAMistral AI
Categorychatchatchat
Charge typePay As You GoFreePay As You Go
Released———
Description
SummaryGemini 4 Carbon is Google's frontier multimodal reasoning model, designed for advanced coding, agentic workflows, and complex multi-step problem solving. It supports text, image, audio, video, and file inputs, enabling integrated reasoning and analysis across diverse modalities. Optimized for long-horizon execution and demanding professional workloads, Gemini 4 Carbon supports up to 1M output tokens through continuation and configurable reasoning effort levels (low, medium, high) to balance reasoning depth, latency, and cost. It is well suited for autonomous agents, software engineering, multimodal analysis, and large-scale workflow automation.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.Codestral is Mistral's cutting-edge language model for coding, released in late July 2025. It is purpose-built for low-latency, high-frequency developer workflows, excelling at tasks such as fill-in-the-middle (FIM) code completion, code correction, and test generation. Optimized for responsiveness and precision, Codestral is well suited for real-time coding assistance, IDE integration, and automated development pipelines where speed and accuracy are critical.