Open-weight GPT model for self-hosted reasoning and instruction-following workloads
Specification
most-agreed values
Context
131K
Max output
33K
Released
2025-08-05
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.07
Output
$0.30
Cache read
$0.035
Quality
3 benchmarks · 1 source
Composite
37th
percentile of 334 scored models
Rank
212 / 334
±0.10 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
14.8
points between effort settings
reasoning31st2/4 bench
math51st1/7 bench
By reasoning effort
every setting placed on the same scale as the leaderboard
| Setting | Composite | Benchmarks behind it |
|---|---|---|
| medium | 48th | 3 |
| low | 32nd | 3 |
| high | 32nd | 3 |
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3 ±5.9 | medium | Epoch AI | 2026-08-27 |
| ↳ | 50.8 ±6.7 | high | Epoch AI | 2026-08-27 |
| ↳ | 40.3 ±6.3 | low | Epoch AI | 2026-08-27 |
| Settings differ by 25.0 points on this benchmark. | ||||
| GPQA diamond | 60.8 ±2.5 | medium | Epoch AI | 2026-08-27 |
| ↳ | 53.2 ±2.5 | low | Epoch AI | 2026-08-27 |
| ↳ | 46.0 ±3.6 | high | Epoch AI | 2026-08-06 |
| Settings differ by 14.8 points on this benchmark. | ||||
| Chess Puzzles | 4.0 ±2.0 | medium | Epoch AI | 2026-08-27 |
| ↳ | 2.0 ±1.4 | low | Epoch AI | 2026-08-27 |
| ↳ | 0.0 ±0.0 | high | Epoch AI | 2026-08-27 |
| Settings differ by 4.0 points on this benchmark. | ||||
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 5 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| ollama · 120b-cloud | free | free | — | — | 131K | 131K | — |
| ollama · 20b-cloud | free | free | — | — | 131K | 131K | — |
| pendra · 120b | free | free | — | — | 131K | 33K | — |
| ollama-cloud · 20b | $0.07 | $0.30 | $0.035 | — | 131K | 33K | — |
| ollama-cloud · 120b | $0.15 | $0.60 | $0.014 | — | 131K | 33K | — |
Other listings of this model
this page is one of them
The corpus files gpt-oss:20b under several keys. This page is the gpt-oss listing. The full record — every host, every price and every benchmark score — is on the main gpt-oss:20b page.
More from OpenAI
most-hosted first