gpt-oss-20b API pricing: 11 providers compared

Where we are not the cheapest, the table says so: 1 provider charges less for output (PrimeIntellect). Sorted by input price, then output. One row per provider: its cheapest listed endpoint for openai/gpt-oss-20b. Other providers' prices are their public list prices as reported by the OpenRouter endpoints API; Synap's are read from /v1/models, which is what the gateway bills. Index generated 2026-11-11 18:07 UTC. Data: price-index.json (CC BY 4.0).

Synap batch price (jobs through /v1/batches, 24-hour window): $0.0085 input / $0.0425 output per 1M tokens. Batch and cached-input discounts from other providers are not in this table.

What is gpt-oss-20b good for?

• High-volume work: classification, extraction, routing
• Tool calling and JSON output
• Chat and summaries over long inputs

Model details, limits and data policy: gpt-oss-20b API.

Call it on Synap

curl https://synap402.com/v1/chat/completions \
  -H "Authorization: Bearer $SYNAP_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-oss-20b", "messages": [{"role": "user", "content": "Hello"}]}'

OpenAI-compatible: any OpenAI SDK works with base_url=https://synap402.com/v1. Tool setup: guides.

Other models in the index

• Qwen3 Coder 30B API pricing
• Gemma 4 26B API pricing
• Qwen3 VL 30B API pricing
• Qwen3 30B A3B Instruct 2507 API pricing
• Muse Glimmer 30B API pricing
• All models, compared
• Cheapest LLM API