gpt-oss-20b
OpenAI released 2025-08-05
$0.055 per million tokens, balanced
Solar Pro 4 is both better and cheaper.
It scores +26.4 higher and costs 5% less ($0.052/M against $0.055/M) on this workload — and it does everything this model does.
Ling-3.0-flash is cheaper still (43% less) but drops 131K → 33K max output.
Every price dimension
| Input | $0.030/M |
|---|---|
| Output | $0.130/M |
| Cached input | $0.030/M |
Independent scores
| Intelligence | 15.2 |
|---|---|
| Coding | 20.7 |
| Agentic | 3.1 |
| LMArena Elo | 1287.8default |
Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.
Capability
| Context window | 131K |
|---|---|
| Max output | 131K |
| Input modes | text |
| Tool use | yes |
| Reasoning | always on |
| Knowledge cutoff | 2024-06-30 |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-08-24 |
| Quality data | verified |
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...