Compare / head-to-head
Claude Sonnet 5.5vs
Qwen3 4B Instruct 2507
Claude Sonnet 5.5 leads 2 of 2 shared benchmarks. Claude Sonnet 5.5 has the larger context window (1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
2 – 0
Claude Sonnet 5.5 leads
Cheaper per token
—
list price, input + output
Larger context
Claude Sonnet 5.5
1M tokens
Providers
2 providers
Anthropic · Qwen
Claude Sonnet 5.5
Anthropic · released 2026-09-28
textvision
Qwen3 4B Instruct 2507
Qwen · released 2025-08-06
text
- Context
- 262K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 18 · 7 core
Quality
Benchmark matrix
2 shared · 27 only Claude Sonnet 5.5 · 16 only Qwen3 4B Instruct 2507| Benchmark | Claude Sonnet 5.5 | Qwen3 4B Instruct 2507 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 2 | ||||
| GPQA Diamond | 95.6% ↗epoch run | 45.8% ↗epoch run | +49.8 pt | Claude Sonnet 5.5 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 100% ↗epoch run | 52.2% ↗epoch run | +47.8 pt | Claude Sonnet 5.5 |
| Only Claude Sonnet 5.5 reports · 27 | ||||
| LMArena Elo | 1471 ↗ | not reported | — | — |
| APEX-Agents (Mercor) | 75.5% ↗ | not reported | — | — |
| Chartography (no tools) | 61.6% ↗ | not reported | — | — |
| CursorBench 4.0 | 55.5% ↗ | not reported | — | — |
| FrontierCode v1.1 (Main) | 46.2% ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 80.5% ↗epoch run | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 88.8% ↗epoch run | not reported | — | — |
| FrontierSWE V2 (Proximal Labs) | 61.9% ↗ | not reported | — | — |
| Furniture Assembly (Epoch AI run) | 75% ↗epoch run | not reported | — | — |
| Humanity's Last Exam (with tools) | 64.5% ↗ | not reported | — | — |
| LiveBench Agentic Coding (LiveBench) | 56.3% ↗ | not reported | — | — |
| LiveBench Coding (LiveBench) | 91.4% ↗ | not reported | — | — |
| LiveBench Data Analysis (LiveBench) | 78.6% ↗ | not reported | — | — |
| LiveBench Instruction Following (LiveBench) | 70.5% ↗ | not reported | — | — |
| LiveBench Language (LiveBench) | 83.4% ↗ | not reported | — | — |
| LiveBench Mathematics (LiveBench) | 96.7% ↗ | not reported | — | — |
| LiveBench Reasoning (LiveBench) | 91.6% ↗ | not reported | — | — |
| LMArena Agent (LMArena) | 0.1252 ↗ | not reported | — | — |
| LMArena Vision (LMArena) | 1268.3 ↗ | not reported | — | — |
| LMArena WebDev (LMArena) | 1786.3 ↗ | not reported | — | — |
| Mystery Game Puzzles (Epoch AI run) | 65% ↗epoch run | not reported | — | — |
| OSWorld 2.1 (partial) | 80.1% ↗ | not reported | — | — |
| SAGE (Vals AI) | 51.8% ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 75.9% ↗ | not reported | — | — |
| SimpleQA Verified | 46.5% ↗epoch run | not reported | — | — |
| Terminal-Bench 4.0 | 70.6% ↗ | not reported | — | — |
| Terminal-Bench 4.0 (Vals AI) | 64.1% ↗ | not reported | — | — |
| Only Qwen3 4B Instruct 2507 reports · 16 | ||||
| MMLU-Pro | not reported | 69.6% ↗ | — | — |
| Aider-Polyglot | not reported | 12.9% ↗ | — | — |
| AIME25 | not reported | 47.4% ↗ | — | — |
| Arena-Hard v2 | not reported | 43.4% ↗ | — | — |
| BFCL-v3 | not reported | 61.9% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 4% ↗epoch run | — | — |
| GPQA | not reported | 62% ↗ | — | — |
| HMMT25 | not reported | 31% ↗ | — | — |
| IFEval | not reported | 83.4% ↗ | — | — |
| LiveBench 20241125 | not reported | 63% ↗ | — | — |
| LiveCodeBench v6 (25.02-25.05) | not reported | 35.1% ↗ | — | — |
| MMLU-Redux | not reported | 84.2% ↗ | — | — |
| MultiPL-E | not reported | 76.8% ↗ | — | — |
| SuperGPQA | not reported | 42.8% ↗ | — | — |
| TAU2-Retail | not reported | 40.4% ↗ | — | — |
| ZebraLogic | not reported | 80.2% ↗ | — | — |
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 4B Instruct 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 | Claude Sonnet 5.5 | Qwen3 4B Instruct 2507 | Edge |
|---|---|---|---|
| Context window | 1M tokens | 262K tokens | Claude Sonnet 5.5 |
| Max output | 128K tokens | — | — |
| Input price / 1M | $2 ↗ | — | — |
| Output price / 1M | $10 ↗ | — | — |
| Cached input / 1M | $0.2 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12 | — | — |
| Modalities | text · vision | text | Claude Sonnet 5.5 |
| Released | 2026-09-28 | 2025-08-06 | — |
| Cited benchmark scores | 29 | 18 | — |
Reliability
Provider status
Live from /statusMore matchups
Claude Sonnet 5.5 vs …
Models sharing the most benchmarksMore matchups
Qwen3 4B Instruct 2507 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 4B Instruct 2507 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.