Gemma 4 31B IT vs GPT-OSS 20B
google/gemma-4-31b-it · openai/gpt-oss-20b
As of October 24, 2026, GPT-OSS 20B is the cheaper API on UberLLM: $0.0002 per million input tokens and $0.0012 per million output, versus $0.0367 / $0.1101 for Gemma 4 31B IT — about 99% less on a blended 3:1 token mix. Both are reached through one OpenAI-compatible key and routed to the cheapest healthy seller, with UberLLM's 10% fee already included, so the price you see is the price you pay. Choose GPT-OSS 20B to cut cost; choose Gemma 4 31B IT when context size wins — Gemma 4 31B IT carries the larger context window (262,144 vs 131,072 tokens), which can matter for long-document work.
Gemma 4 31B IT vs GPT-OSS 20B, side by side
| Spec | Gemma 4 31B IT | GPT-OSS 20B |
|---|---|---|
| Price · input / 1M | $0.0367 | $0.0002 |
| Price · output / 1M | $0.1101 | $0.0012 |
| Context length | 262,144 | 131,072 |
| Max output tokens | 131,072 | 32,768 |
| Input modalities | text, image | text |
| Output modalities | text | text |
| Tools / functions | Yes | Yes |
| Vision (image input) | Yes | — |
| Cheapest seller | UberLLM | UberLLM |
| % off list | 69% off | 99% off |
Live prices via UberLLM, routed to the cheapest healthy seller. The price shown includes UberLLM’s 10% fee. Updated October 24, 2026.
Gemma 4 31B IT vs GPT-OSS 20B — FAQ
Is Gemma 4 31B IT or GPT-OSS 20B cheaper?
GPT-OSS 20B is cheaper on a blended token mix, by about 99%. UberLLM routes every request to the cheapest healthy seller and bills the price shown, with its 10% fee included.
What's the price difference between Gemma 4 31B IT and GPT-OSS 20B?
Gemma 4 31B IT: $0.0367 in / $0.1101 out per 1M tokens. GPT-OSS 20B: $0.0002 in / $0.0012 out per 1M tokens. That's roughly a 99% blended difference. Prices are live and include UberLLM's 10% fee.
Can I use both Gemma 4 31B IT and GPT-OSS 20B with one API?
Yes. Both are served through a single OpenAI-compatible endpoint — set the base URL to https://api.uberllm.dev/v1 and your UberLLM key, then switch between them by changing the model id. Anthropic SDKs work via https://api.uberllm.dev/v1/messages.