Pairwise benchmark snapshot · Aug 17, 2026noindex (thin pair)
This page compares Claude Opus 4.6 (Anthropic) and OpenAI: gpt-oss-20b (free) (OpenAI) using the latest snapshots we have as of Aug 17, 2026. OpenAI: gpt-oss-20b (free) wins on output price at $0/1M. Claude Opus 4.6 has the larger context window (1M tokens). Every cell below is a dated snapshot from a named public source — we do not invent missing scores, and we do not run SWE-bench ourselves.
Anthropic
Claude Opus 4.6Follow-on Opus release. Slightly behind 4.5 on official SWE-bench bash-only in the last published sweep; stronger long-horizon agent traces.
OpenAI
OpenAI: gpt-oss-20b (free)Auto-discovered from OpenRouter (openai/gpt-oss-20b:free). Preview until a second source matches.
thinking_budgetReasoning models scale test-time compute by generating hidden chain-of-thought (CoT) tokens. These tokens are billed at standard output rates and directly expand latency (TTFT) in exchange for higher GPQA Diamond and SWE-bench Verified problem resolve rates.
Official evaluation benchmark budget for SWE-bench and LiveBench.
"thinking": {
"type": "enabled",
"budget_tokens": 16384
}"reasoning": {
"effort": "medium",
"max_tokens": 3200
}Instant pairwise benchmark recalculation across 400+ LLMs
Category wins across intelligence, speed, and pricing efficiency.
Preference Elo, Coding proficiency, SWE-bench & LiveBench accuracy
Generation throughput and time to first token responsiveness
Cost per million tokens and max context window length
Each spoke is a skill. Farther from the center is better (0–100th percentile). Tap any dot to inspect details.
No shared data
No shared data
$25.00/1M cheaper
Need SWE-bench on both sides.
Lower output list price ($0/1M).
Need TTFT on both sides.
Larger window (1M).
Claude Opus 4.6 is the side marked multimodal in the catalog.
| Target Workload | Recommended Pick | Evaluation Rationale |
|---|---|---|
| Repo / coding agents | insufficient data | Need SWE-bench on both sides. |
| High-volume chat | OpenAI: gpt-oss-20b (free) | Lower output list price ($0/1M). |
| Voice / low-latency UI | insufficient data | Need TTFT on both sides. |
| Long-document RAG | Claude Opus 4.6 | Larger window (1M). |
| Screenshots / vision | Claude Opus 4.6 | Claude Opus 4.6 is the side marked multimodal in the catalog. |
No comments posted on this matchup yet. Be the first to share an evaluation note!
The real PNG is generated at /compare/claude-opus-4-6-vs-gpt-oss-20b-free/opengraph-image for crawlers.
Verified head-to-head card