Amazon: Nova Micro 1.0 vs OpenAI: gpt-oss-20b
On input price, OpenAI: gpt-oss-20b is ~1.9× cheaper. Verified 2026-10-21.
Amazon: Nova Micro 1.0 $0.035/$0.140vsOpenAI: gpt-oss-20b $0.018/$0.090
Verdict
On blended price (3:1 input:output), OpenAI: gpt-oss-20b is ~1.7× cheaper ($0.018/$0.090 vs $0.035/$0.140 per 1M tokens).
Pricing
Per-million-token list prices, verified 2026-10-21. Green marks the better value.
| Metric | Amazon: Nova Micro 1.0 | OpenAI: gpt-oss-20b |
|---|---|---|
| Input / 1M | $0.035 | $0.018 |
| Output / 1M | $0.140 | $0.090 |
| Cache read / 1M | — | $0.0090 |
| Blended (3:1) | $0.061/1M | $0.036/1M |
| Estimated monthly cost | Amazon: Nova Micro 1.0 | OpenAI: gpt-oss-20b |
|---|---|---|
| Chatbot10,000 reqs · 800/400 tok | $0.84 | $0.50 |
| Coding agent5,000 reqs · 8000/2000 tok | $2.80 | $1.62 |
| RAG pipeline30,000 reqs · 4000/500 tok | $6.30 | $3.51 |
Plug in your own numbers in the calculator, or paste a real bill into the Bill Shredder.
Specifications & capabilities
| Metric | Amazon: Nova Micro 1.0 | OpenAI: gpt-oss-20b |
|---|---|---|
| Context window | 128K | 131K |
| Max output | 5K | 33K |
| Vision input | – | – |
| Tool calling | ✓ | ✓ |
| Reasoning | – | ✓ |
Provider
Infrastructure metrics for the organizations behind these models.
| Metric | Amazon Nova | OpenAI |
|---|---|---|
| Listed endpoints | 4 | 12 |
| Reliability | 99.7% | 99.9% |
| Avg latency | 380 ms | 450 ms |
| Pricing tiers | standard | standard, batch, provisioned |
Full details: Amazon: Nova Micro 1.0 pricing · OpenAI: gpt-oss-20b pricing