gpt-oss or Qwen: which one to run on your own server

Both families install through Ollama and run in a closed environment with no cloud. gpt-oss comes in two sizes, 20B and 120B, entirely under Apache 2.0 with unconditional commercial use, but it needs a GPU. Qwen starts at 0.6B, runs on a CPU and is listed in the catalog as supporting Russian, though the larger Qwen3.8 models carry their own license and conditional commercial terms. On dates, the catalog has a single gpt-oss release from August 2025, while Qwen versions continue through August 2026.

Comparison based on catalog data

Parametergpt-ossQwen
CategoryTextText, Image + text
DeveloperOpenAI, USAAlibaba, China
ReleasesAug 2025 – Aug 2025Aug 2023 – Aug 2026
Sizes20B, 120B0,6B – 2,4T-A95B
Hardware1 GPULaptop, 1 GPU, Cluster
Commercial useCommercial use allowedCommercial use with conditions
LicenseApache 2.0Apache 2.0 (most versions); the larger Qwen3.8 models have their own license
RussianNot statedSupported
OllamaYesYes
Without GPUNoYes
Tasks
  • AI agent that calls internal systems
  • Answers based on internal policies
  • Drafts of emails and reports
  • Chatbot and knowledge-base assistant
  • Replies to emails and customer requests
  • Document parsing and classification
  • Product description generation

Choose gpt-oss if

  • You want plain Apache 2.0 with no strings attached to commercial use
  • You are building a corporate agent that calls internal systems
  • You have a GPU: the smaller gpt-oss-20b fits on one card
gpt-oss

Choose Qwen if

  • Your assistant must answer in Russian, which Qwen lists and gpt-oss does not
  • No GPU available: Qwen starts at 0.6B and runs on a CPU
  • You want current releases and sizes all the way up to 2.4T-A95B
Qwen

Other comparisons

Need a model for your task?

An open model can run on your own server: data stays in-house, there is no per-request fee, and the model can be fine-tuned on your documents.

  1. SelectThe model and size for your task and hardware budget
  2. DeployOn your server or in a closed network, with an API
  3. Fine-tuneOn your data, or connect a knowledge base
  4. IntegrateInto your CRM, ERP, bot, website or team chat
Discuss deployment