GPT-6 Astra vs gpt-oss-20b
GPT-6 Astra (OpenAI) and gpt-oss-20b (OpenAI) compared on benchmarks, pricing, context window, and use-case rankings.
- GPT-6 Astra leads on overall intelligence (ECI 169 vs 138).
- gpt-oss-20b is cheaper on input tokens — $0.02 vs $10.00 per 1M.
- GPT-6 Astra has a larger context window (1.1M vs 131K tokens).
- GPT-6 Astra ranks higher for coding (93th vs 10th percentile).
| Attribute | GPT-6 Astra | gpt-oss-20b |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 169 | 138 |
| Coding | 93 | 10 |
| Math | 91 | 4 |
| Reasoning & Knowledge | 98 | 6 |
| Agentic & Tools | 90 | 6 |
| Specifications | ||
| Developer | OpenAI | OpenAI |
| Family | GPT | GPT |
| Released | Sep 3, 2026 | Aug 5, 2025 |
| Parameters | — | 20.9B |
| Availability | API access | Open weights (unrestricted) |
| Context window | 1.1M | 131K |
| Price — $/M input | $10.00 | $0.02 |
| Price — $/M output | $50.00 | $0.09 |
| Inputs | file, image, text | text |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 100% | 65% |
| APEX | 65% | — |
| ARC-AGI | 99% | — |
| ARC-AGI-2 | 95% | — |
| GPQA Diamond | 96% | 61% |
| GSO (code optimization) | 79% | — |
| Humanity's Last Exam | 55% | 5% |
| SciCode | 56% | 39% |
| SimpleQA Verified | 76% | — |
| WebDev Arena | 1800 | — |
| WeirdML | 94% | 41% |
| LiveCodeBench | — | 65% |
| MMLU-Pro | — | 72% |
| Terminal-Bench | — | 3% |
| τ²-bench | — | 50% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Is GPT-6 Astra better than gpt-oss-20b?
On Epoch AI's Capabilities Index, GPT-6 Astra scores higher (169) than gpt-oss-20b (138). The right pick depends on your task — compare their coding, math, and reasoning scores in the table above.
Which is cheaper, GPT-6 Astra or gpt-oss-20b?
gpt-oss-20b is cheaper on input tokens at $0.02 per million, versus $10.00 (representative OpenRouter pricing).
Which has a larger context window, GPT-6 Astra or gpt-oss-20b?
GPT-6 Astra supports up to 1.1M tokens, compared with 131K for the other.
Which is better for coding, GPT-6 Astra or gpt-oss-20b?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, GPT-6 Astra ranks higher — 93th vs 10th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.