ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

gpt-oss-20bvsLlama 4 Scout

gpt-oss-20b leads 1 of 1 shared benchmark. Llama 4 Scout has the larger context window (10M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
1 – 0
gpt-oss-20b leads
Cheaper per token
list price, input + output
Larger context
Llama 4 Scout
10M tokens
Provider uptime (30d)
94.2446% · —
OpenAI · Meta
gpt-oss-20b
OpenAI · released 2025-08-05
text
Context
131K
Max out
131K
Input /1M
Output /1M
Cached /1M
Scores
6 · 5 core
Llama 4 Scout
Meta · released 2025-04-05
textvision
Context
10M
Max out
Input /1M
Output /1M
Cached /1M
Scores
2 · 2 core
Quality

Benchmark matrix

Benchmarkgpt-oss-20bLlama 4 ScoutΔEdge
Reported by both · 1
GPQA Diamond71.5% 57.2% +14.3 ptgpt-oss-20b
Only gpt-oss-20b reports · 4
SWE-bench Verified60.7% not reported
AIME 202589.2% matharenanot reported
LMArena Elo1287.3 not reported
OTIS Mock AIME 2024-2025 (Epoch AI run)65.3% epochnot reported
Only Llama 4 Scout reports · 1
MMLU-Pronot reported74.3%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is gpt-oss-20b minus Llama 4 Scout in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

Specgpt-oss-20bLlama 4 ScoutEdge
Context window131K tokens10M tokensLlama 4 Scout
Max output131K tokens
Input price / 1M
Output price / 1M
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
Modalitiestexttext · visionLlama 4 Scout
Released2025-08-052025-04-05
Cited benchmark scores62
Reliability

Provider status

All providers
More matchups

gpt-oss-20b vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run gpt-oss-20b and Llama 4 Scout on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.