A list of recommended open models for common use cases.

​
Which open models should I use?

Actually, there’s no single right answer! Here’s a curated list based on Novita internal testing, community feedback, and external benchmarks. We recommend using it as a starting point, and we will update it regularly as new models emerge.
  • Model sizes are marked as Small, Medium, or Large
  • For best latency, use small or medium models. For best quality, use large models or fine-tune medium or small models.
  • You can explore all models in the Novita Model Library
Use CasesRecommended Models
Code generation & reasoningDeepseek-r1-0528, Deepseek-v3-0324,Qwen3-Coder-480B-A35B-Instruct,Qwen3-235B-A22B-Instruct-2507_(Large) _Kimi-K2-Instruct,GLM-4.5(Medium)
General reasoning & planningDeepseek-r1-0528, Deepseek-v3-0324 (Large)
Qwen-2.5-72b-instruct, Llama-3.3-70b-instruct (Medium)
Function calling & tool useQwen3-235b-a22b-fp8 (Large)
Qwen 3 Family Models (Large/Medium/Small)
Long context & summarizationLlama-4-maverick-17b-128e-instruct-fp8 (Large)
Llama-4-scout-17b-16e-instruct (Medium)
Vision & document understandingLlama-4-maverick-17b-128e-instruct-fp8 (Large)
Qwen2.5-vl-72b-instruct, Llama-4-scout-17b-16e-instruct (Medium)
Low-latency NLU & extractionLlama-3.1-8b-instruct, Llama-3.2-3b-instruct, Llama-3.2-1b-instruct, Qwen3-8b-fp8 (Small)
Last updated: August 1, 2025