gpt-oss-20b
by OpenAI · openai/gpt-oss-20b · #16 cheapest of 457 paid models
Prices updated Nov 18, 2026, 4:47 PM UTC · refreshed nightly
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5Gemini 3.6 Flash
Pricing
Input
$0.018
per 1M tokens
Output
$0.09
per 1M tokens
Blended (3:1)
$0.036
3 input : 1 output
Cache read
$0.0090
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for gpt-oss-20b with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.036 | $0.001926 | — |
| 50% | $0.033 | $0.001476 | −23% |
| 90% | $0.03 | $0.001116 | −42% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates for this provider’s batch API. Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $3.60e-5 | $0.036 |
| Document summary | 8,000 | 1,000 | $0.000234 | $0.234 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.00072 | $0.72 |
| Long-context analysis | 150,000 | 5,000 | $0.00315 | $3.15 |
Performance
Intelligence Index
—
AA composite quality
Coding Index
20.7
Math Index
—
Agentic Index
1.2
Output speed
—
median
Time to first token
—
median
Time to first answer
—
after reasoning tokens
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 843 |
| Data viz | 933 |
Specs
Context window
131K
tokens
Max output
118K
tokens
Input modalities
text
Tokenizer
GPT