Compare / head-to-head
GPT-5.2-CodexvsGPT-6.1 Sol
GPT-6.1 Sol leads 7 of 8 shared benchmarks. GPT-6.1 Sol is 1.3x cheaper per token. GPT-6.1 Sol has the larger context window (1.1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 7
GPT-6.1 Sol leads
Cheaper per token
GPT-6.1 Sol
1.3x cheaper, input + output
Larger context
GPT-6.1 Sol
1.1M tokens
Providers
OpenAI
same provider
GPT-5.2-Codex
OpenAI · released 2025-12-18
textvision
GPT-6.1 Sol
OpenAI · released 2026-09-29
textvision
Quality
Benchmark matrix
8 shared · 5 only GPT-5.2-Codex · 16 only GPT-6.1 Sol| Benchmark | GPT-5.2-Codex | GPT-6.1 Sol | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 8 | ||||
| LiveBench Agentic Coding (LiveBench) | 49.4% ↗ | 56.8% ↗ | -7.4 pt | GPT-6.1 Sol |
| LiveBench Coding (LiveBench) | 83.6% ↗ | 80.7% ↗ | +2.9 pt | GPT-5.2-Codex |
| LiveBench Data Analysis (LiveBench) | 78.2% ↗ | 82.7% ↗ | -4.5 pt | GPT-6.1 Sol |
| LiveBench Instruction Following (LiveBench) | 66.4% ↗ | 74.2% ↗ | -7.8 pt | GPT-6.1 Sol |
| LiveBench Language (LiveBench) | 73.7% ↗ | 90.1% ↗ | -16.4 pt | GPT-6.1 Sol |
| LiveBench Mathematics (LiveBench) | 88.8% ↗ | 96.8% ↗ | -8 pt | GPT-6.1 Sol |
| LiveBench Reasoning (LiveBench) | 77.7% ↗ | 92.6% ↗ | -14.9 pt | GPT-6.1 Sol |
| LMArena WebDev (LMArena) | 1338.9 ↗ | 1757.8 ↗ | -418.9 | GPT-6.1 Sol |
| Only GPT-5.2-Codex reports · 5 | ||||
| Cybersecurity Capture The Flag Challenges | 67.4% ↗ | not reported | — | — |
| OSWorld-Verified | 38.2% ↗ | not reported | — | — |
| SWE-Bench Pro | 56.4% ↗ | not reported | — | — |
| SWE-Lancer IC Diamond | 76% ↗ | not reported | — | — |
| Terminal-Bench 2.0 | 64% ↗ | not reported | — | — |
| Only GPT-6.1 Sol reports · 16 | ||||
| GPQA Diamond | not reported | 95.4% ↗epoch run | — | — |
| LMArena Elo | not reported | 1483.2 ↗ | — | — |
| APEX-Agents (Mercor) | not reported | 60% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 61% ↗epoch run | — | — |
| EBR-bench (Epoch AI run) | not reported | 54.3% ↗epoch run | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 100% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 93.7% ↗epoch run | — | — |
| Furniture Assembly (Epoch AI run) | not reported | 80% ↗epoch run | — | — |
| LMArena Agent (LMArena) | not reported | 0.1123 ↗ | — | — |
| LMArena Vision (LMArena) | not reported | 1290.6 ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 80% ↗epoch run | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 100% ↗epoch run | — | — |
| SAGE (Vals AI) | not reported | 46.5% ↗ | — | — |
| SimpleBench (SimpleBench) | not reported | 82.9% ↗ | — | — |
| SimpleQA Verified | not reported | 73.9% ↗epoch run | — | — |
| Terminal-Bench 4.0 (Vals AI) | not reported | 55.1% ↗ | — | — |
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-5.2-Codex minus GPT-6.1 Sol 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
Official pricing pages and model cards| Spec | GPT-5.2-Codex | GPT-6.1 Sol | Edge |
|---|---|---|---|
| Context window | 400K tokens | 1.1M tokens | GPT-6.1 Sol |
| Max output | 128K tokens | 128K tokens | Tie |
| Input price / 1M | $1.75 ↗ | $2 ↗ | GPT-5.2-Codex |
| Output price / 1M | $14 ↗ | $10 ↗ | GPT-6.1 Sol |
| Cached input / 1M | $0.175 ↗ | $0.1 ↗ | GPT-6.1 Sol |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $15.75 | $12 | GPT-6.1 Sol |
| Modalities | text · vision | text · vision | Tie |
| Released | 2025-12-18 | 2026-09-29 | — |
| Cited benchmark scores | 13 | 24 | — |
Reliability
OpenAI status
Live from /statusMore matchups
GPT-5.2-Codex vs …
Models sharing the most benchmarksMore matchups
GPT-6.1 Sol vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GPT-5.2-Codex and GPT-6.1 Sol on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.