Compare / head-to-head
Claude Sonnet 5.5vs
Qwen3.8-Max
Claude Sonnet 5.5 leads 18 of 22 shared benchmarks. Qwen3.8-Max is 1.5x cheaper per token. Both offer a 1M-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
18 – 4
Claude Sonnet 5.5 leads
Cheaper per token
Qwen3.8-Max
1.5x cheaper, input + output
Larger context
Tie
both 1M tokens
Providers
2 providers
Anthropic · Qwen
Claude Sonnet 5.5
Anthropic · released 2026-09-28
textvision
Qwen3.8-Max
Qwen · released 2026-08-02
textvision
Quality
Benchmark matrix
22 shared · 7 only Claude Sonnet 5.5 · 5 only Qwen3.8-Max| Benchmark | Claude Sonnet 5.5 | Qwen3.8-Max | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 22 | ||||
| GPQA Diamond | 95.6% ↗epoch run | 92.6% ↗ | +3 pt | Claude Sonnet 5.5 |
| LMArena Elo | 1471 ↗ | 1480.6 ↗ | -9.6 | Qwen3.8-Max |
| APEX-Agents (Mercor) | 75.5% ↗ | 63.3% ↗ | +12.2 pt | Claude Sonnet 5.5 |
| FrontierMath Tier 4 v2 (Epoch AI run) | 80.5% ↗epoch run | 46.3% ↗epoch run | +34.2 pt | Claude Sonnet 5.5 |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 88.8% ↗epoch run | 74.7% ↗epoch run | +14.1 pt | Claude Sonnet 5.5 |
| FrontierSWE V2 (Proximal Labs) | 61.9% ↗ | 17.8% ↗ | +44.1 pt | Claude Sonnet 5.5 |
| Furniture Assembly (Epoch AI run) | 75% ↗epoch run | 20% ↗epoch run | +55 pt | Claude Sonnet 5.5 |
| LiveBench Agentic Coding (LiveBench) | 56.3% ↗ | 64.6% ↗ | -8.3 pt | Qwen3.8-Max |
| LiveBench Coding (LiveBench) | 91.4% ↗ | 72.9% ↗ | +18.5 pt | Claude Sonnet 5.5 |
| LiveBench Data Analysis (LiveBench) | 78.6% ↗ | 78.4% ↗ | +0.2 pt | Claude Sonnet 5.5 |
| LiveBench Instruction Following (LiveBench) | 70.5% ↗ | 74.1% ↗ | -3.6 pt | Qwen3.8-Max |
| LiveBench Language (LiveBench) | 83.4% ↗ | 79.7% ↗ | +3.7 pt | Claude Sonnet 5.5 |
| LiveBench Mathematics (LiveBench) | 96.7% ↗ | 91.3% ↗ | +5.4 pt | Claude Sonnet 5.5 |
| LiveBench Reasoning (LiveBench) | 91.6% ↗ | 88.2% ↗ | +3.4 pt | Claude Sonnet 5.5 |
| LMArena Agent (LMArena) | 0.1252 ↗ | 0.0251 ↗ | +0.1 | Claude Sonnet 5.5 |
| LMArena Vision (LMArena) | 1268.3 ↗ | 1301.2 ↗ | -32.9 | Qwen3.8-Max |
| LMArena WebDev (LMArena) | 1786.3 ↗ | 1671.2 ↗ | +115.1 | Claude Sonnet 5.5 |
| Mystery Game Puzzles (Epoch AI run) | 65% ↗epoch run | 38% ↗epoch run | +27 pt | Claude Sonnet 5.5 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 100% ↗epoch run | 99.4% ↗epoch run | +0.6 pt | Claude Sonnet 5.5 |
| SAGE (Vals AI) | 51.8% ↗ | 51.3% ↗ | +0.5 pt | Claude Sonnet 5.5 |
| SimpleQA Verified | 46.5% ↗epoch run | 45.8% ↗epoch run | +0.7 pt | Claude Sonnet 5.5 |
| Terminal-Bench 4.0 (Vals AI) | 64.1% ↗ | 34.3% ↗ | +29.8 pt | Claude Sonnet 5.5 |
| Only Claude Sonnet 5.5 reports · 7 | ||||
| Chartography (no tools) | 61.6% ↗ | not reported | — | — |
| CursorBench 4.0 | 55.5% ↗ | not reported | — | — |
| FrontierCode v1.1 (Main) | 46.2% ↗ | not reported | — | — |
| Humanity's Last Exam (with tools) | 64.5% ↗ | not reported | — | — |
| OSWorld 2.1 (partial) | 80.1% ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 75.9% ↗ | not reported | — | — |
| Terminal-Bench 4.0 | 70.6% ↗ | not reported | — | — |
| Only Qwen3.8-Max reports · 5 | ||||
| Chess Puzzles (Epoch AI run) | not reported | 40% ↗epoch run | — | — |
| DeepSWE v1.1 (Datacurve) | not reported | 57.5% ↗ | — | — |
| SWE-Bench Pro | not reported | 67.7% ↗ | — | — |
| tau2-bench Banking Knowledge (Sierra) | not reported | 55.2% ↗ | — | — |
| Terminal-Bench 2.1 | not reported | 86.6% ↗ | — | — |
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 Claude Sonnet 5.5 minus Qwen3.8-Max 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 | Claude Sonnet 5.5 | Qwen3.8-Max | Edge |
|---|---|---|---|
| Context window | 1M tokens | 1M tokens | Tie |
| Max output | 128K tokens | 131K tokens | Qwen3.8-Max |
| Input price / 1M | $2 ↗ | $2 ↗ | Tie |
| Output price / 1M | $10 ↗ | $6 ↗ | Qwen3.8-Max |
| Cached input / 1M | $0.2 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12 | $8 | Qwen3.8-Max |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-09-28 | 2026-08-02 | — |
| Cited benchmark scores | 29 | 34 | — |
Reliability
Provider status
Live from /statusMore matchups
Claude Sonnet 5.5 vs …
Models sharing the most benchmarksMore matchups
Qwen3.8-Max vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Claude Sonnet 5.5 and Qwen3.8-Max on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.