Skip to content

Nemotron 3.5 Super VL vs Mistral Medium 3.5

Nemotron 3.5 Super VL is from NVIDIA, released October 15, 2026. Mistral Medium 3.5 is from Mistral AI, released April 28, 2026.

  • LMArena Elo1416 vs 1426. Mistral Medium 3.5 is 0.7% ahead on LMArena Elo.
  • Context window1M vs 256k. Nemotron 3.5 Super VL takes 291% more context in one request.

Prices are the vendors’ published API rates; Elo is LMArena’s text board. Each figure is dated on its model page.

Interactive Comparison

Loading chart...
Free — open weights, self-hosted (no first-party API pricing)
$1.50 / $7.50 per million tokens
Identity
DeveloperNVIDIAMistral AI
ReleasedOct 2026Apr 2026
StatusGAGA
LicenceOpenMDW License Agreement 1.1Open weights (Modified MIT)
Self-hostableYes — open weights (BF16 and NVFP4 releases)Yes (open weights)
Cost
Blended $/1M tokensinput × 0.75 + output × 0.25—$3.000 / 1M tokens
Input priceSelf-hosted — no first-party API price$1.500 / 1M tokens
Output priceSelf-hosted — no first-party API price$7.500 / 1M tokens
Cached input—Not published
Batch discount—50% off (Batch API)
Free tier—Free tier via La Plateforme rate limits
Capacity
Context window1M tokens256k tokens
Max output—Not published
Long-context surcharge—Not published
Capability
Vision inYes — images and videoYes
Audio inNoNo
Function calling—Yes
Structured output—Yes
Extended reasoning—Yes (configurable reasoning effort)
Web search—No
Code execution—No
Access
APINot offered first-partyYes
Chat app—Yes
Cloud marketplacesNVIDIA NIMNot published
Fine-tuningYes — open weights permit itNot published
Measured quality
LMArena Elo1416 (checked Oct 2026)1426 (checked Oct 2026)
LMArena rankRank 138Rank 117
Elo per dollarLMArena Elo ÷ blended $/1M tokens—475

Add a third model to this comparison