ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GPT-5vsGPT-5.1

GPT-5 leads 4 of 5 shared benchmarks. Both list the same combined token price. Both offer a 400K-token context window.

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

Shared benchmarks
4 – 1
GPT-5 leads
Cheaper per token
Tie
list price, input + output
Larger context
Tie
both 400K tokens
Provider uptime (30d)
94.2446%
OpenAI
GPT-5
OpenAI · released 2025-08-07
textvision
Context
400K
Max out
128K
Input /1M
$1.25
Output /1M
$10
Cached /1M
$0.125
Scores
9 · 8 core
GPT-5.1
OpenAI · released 2025-11-13
textvision
Context
400K
Max out
128K
Input /1M
$1.25
Output /1M
$10
Cached /1M
$0.125
Scores
6 · 6 core
Quality

Benchmark matrix

BenchmarkGPT-5GPT-5.1ΔEdge
Reported by both · 5
GPQA Diamond86.2% epoch87.6% epoch-1.4 ptGPT-5.1
AIME 202594.6% 94.2% matharena+0.4 ptGPT-5
OTIS Mock AIME 2024-2025 (Epoch AI run)91.4% epoch88.6% epoch+2.8 ptGPT-5
SimpleQA Verified50.1% epoch48% epoch+2.1 ptGPT-5
SWE-bench Verified (Epoch AI run)73.6% epoch68% epoch+5.6 ptGPT-5
Only GPT-5 reports · 3
SWE-bench Verified74.9% not reported
FrontierMath Tier 4 v2 (Epoch AI run)22% epochnot reported
FrontierMath Tiers 1-3 v2 (Epoch AI run)55.4% epochnot reported
Only GPT-5.1 reports · 1
LMArena Elonot reported1422.6
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-5 minus GPT-5.1 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-5GPT-5.1Edge
Context window400K tokens400K tokensTie
Max output128K tokens128K tokensTie
Input price / 1M$1.25 $1.25 Tie
Output price / 1M$10 $10 Tie
Cached input / 1M$0.125 $0.125 Tie
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$11.25$11.25Tie
Modalitiestext · visiontext · visionTie
Released2025-08-072025-11-13
Cited benchmark scores96
Reliability

OpenAI status

All providers
Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GPT-5 and GPT-5.1 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.