Compare / head-to-head
GPT-6 Solvs
Qwen3 235B-A22B Thinking 2507
GPT-6 Sol leads 6 of 6 shared benchmarks. Qwen3 235B-A22B Thinking 2507 is 4.7x cheaper per token. GPT-6 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
6 – 0
GPT-6 Sol leads
Cheaper per token
Qwen3 235B-A22B Thinking 2507
4.7x cheaper, input + output
Larger context
GPT-6 Sol
1.1M tokens
Providers
2 providers
OpenAI · Qwen
GPT-6 Sol
OpenAI · released 2026-09-22
textvision
Qwen3 235B-A22B Thinking 2507
Qwen · released 2025-07-25
text
Quality
Benchmark matrix
6 shared · 21 only GPT-6 Sol · 17 only Qwen3 235B-A22B Thinking 2507| Benchmark | GPT-6 Sol | Qwen3 235B-A22B Thinking 2507 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 6 | ||||
| GPQA Diamond | 94.3% ↗epoch run | 80.1% ↗epoch run | +14.2 pt | GPT-6 Sol |
| LMArena Elo | 1457.2 ↗ | 1400 ↗ | +57.2 | GPT-6 Sol |
| Mystery Game Puzzles (Epoch AI run) | 56% ↗epoch run | 9% ↗epoch run | +47 pt | GPT-6 Sol |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 100% ↗epoch run | 86.7% ↗epoch run | +13.3 pt | GPT-6 Sol |
| SimpleQA Verified | 60.7% ↗epoch run | 40.4% ↗epoch run | +20.3 pt | GPT-6 Sol |
| Vending-Bench 2 (Andon Labs) | 14427.85 ↗ | -11.34 ↗ | +14439.2 | GPT-6 Sol |
| Only GPT-6 Sol reports · 21 | ||||
| Agents' Last Exam V1 | 56.4% ↗ | not reported | — | — |
| APEX-Agents (Mercor) | 54.3% ↗ | not reported | — | — |
| AutomationBench 1.0.6 | 33.2% ↗ | not reported | — | — |
| DeepSWE v1.1 | 68.8% ↗ | not reported | — | — |
| EBR-bench (Epoch AI run) | 53.3% ↗epoch run | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 90% ↗epoch run | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 89.8% ↗epoch run | not reported | — | — |
| Furniture Assembly (Epoch AI run) | 58.3% ↗epoch run | not reported | — | — |
| LiveBench Agentic Coding (LiveBench) | 52.9% ↗ | not reported | — | — |
| LiveBench Coding (LiveBench) | 81.8% ↗ | not reported | — | — |
| LiveBench Data Analysis (LiveBench) | 81.2% ↗ | not reported | — | — |
| LiveBench Instruction Following (LiveBench) | 68.6% ↗ | not reported | — | — |
| LiveBench Language (LiveBench) | 85.3% ↗ | not reported | — | — |
| LiveBench Mathematics (LiveBench) | 96.4% ↗ | not reported | — | — |
| LiveBench Reasoning (LiveBench) | 88.7% ↗ | not reported | — | — |
| LMArena Agent (LMArena) | 0.0971 ↗ | not reported | — | — |
| LMArena WebDev (LMArena) | 1689 ↗ | not reported | — | — |
| OSWorld 2.0 (offline, partial) | 60.5% ↗ | not reported | — | — |
| SAGE (Vals AI) | 44.8% ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 73.1% ↗ | not reported | — | — |
| Terminal-Bench 4.0 (Vals AI) | 44.4% ↗ | not reported | — | — |
| Only Qwen3 235B-A22B Thinking 2507 reports · 17 | ||||
| MMLU-Pro | not reported | 84.4% ↗ | — | — |
| AIME25 | not reported | 92.3% ↗ | — | — |
| Arena-Hard v2 | not reported | 79.7% ↗ | — | — |
| BFCL-v3 | not reported | 71.9% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 12% ↗epoch run | — | — |
| GPQA | not reported | 81.1% ↗ | — | — |
| HLE (text-only) | not reported | 18.2% ↗ | — | — |
| HMMT25 | not reported | 83.9% ↗ | — | — |
| IFEval | not reported | 87.8% ↗ | — | — |
| LiveBench 20241125 | not reported | 78.4% ↗ | — | — |
| LiveCodeBench v6 (25.02-25.05) | not reported | 74.1% ↗ | — | — |
| MMLU-Redux | not reported | 93.8% ↗ | — | — |
| OJBench | not reported | 32.5% ↗ | — | — |
| SuperGPQA | not reported | 64.9% ↗ | — | — |
| TAU2-Airline | not reported | 58% ↗ | — | — |
| TAU2-Retail | not reported | 71.9% ↗ | — | — |
| WeirdML (Håvard Tveit Ihle) | not reported | 41% ↗ | — | — |
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-6 Sol minus Qwen3 235B-A22B Thinking 2507 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-6 Sol | Qwen3 235B-A22B Thinking 2507 | Edge |
|---|---|---|---|
| Context window | 1.1M tokens | 262K tokens | GPT-6 Sol |
| Max output | 128K tokens | — | — |
| Input price / 1M | $2 ↗ | $0.23 ↗ | Qwen3 235B-A22B Thinking 2507 |
| Output price / 1M | $10 ↗ | $2.3 ↗ | Qwen3 235B-A22B Thinking 2507 |
| Cached input / 1M | $0.2 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12 | $2.53 | Qwen3 235B-A22B Thinking 2507 |
| Modalities | text · vision | text | GPT-6 Sol |
| Released | 2026-09-22 | 2025-07-25 | — |
| Cited benchmark scores | 27 | 23 | — |
Reliability
Provider status
Live from /statusMore matchups
GPT-6 Sol vs …
Models sharing the most benchmarksMore matchups
Qwen3 235B-A22B Thinking 2507 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GPT-6 Sol and Qwen3 235B-A22B Thinking 2507 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.