Which Open Models Should I Use?

There’s no single right answer! Here’s a curated list based on Fireworks internal testing, community feedback, and external benchmarks. We recommend you use it as a starting point, and we will update it regularly as new models emerge.

Note:

  • 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 Fireworks Model Library
Use CaseRecommended Models
Code generation & reasoningDeepSeek R1, V3-0324 (Large)
Qwen2.5-32B-Coder (Medium)
Code completion & bug fixingQwen2.5-32B-Coder (Medium)
DeepSeek V2.5 (Medium)
Qwen2.5 0.5–14B (Small)
General reasoning & planningDeepSeek R1, V3-0324 (Large)
Qwen2.5-72B-Instruct (Medium)
Llama 3.3 70B (Medium)
Function calling & tool useQwen2.5-72B-Instruct (Medium)
Long context & summarizationLlama 4 Maverick & Scout (Medium/Large)
Vision & document understandingQwen2.5-32B-VL, 72B-VL (Medium)
Llama 4 Maverick & Scout (Medium/Large)
Qwen2.5 3–7B (Small)
Low-latency NLU & extractionLlama 3.1 8B, 3.2 3B/1B (Small)
Qwen2.5 0.5–7B (Small)

Last updated: April 28, 2025