Compare / head-to-head
Qwen3 4B Instruct 2507vsQwen3-Next 80B-A3B Thinking
Qwen3-Next 80B-A3B Thinking leads 11 of 11 shared benchmarks. Both offer a 262K-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
0 – 11
Qwen3-Next 80B-A3B Thinking leads
Cheaper per token
—
list price, input + output
Larger context
Tie
both 262K tokens
Providers
Qwen
same provider
Qwen3 4B Instruct 2507
Qwen · released 2025-08-06
text
- Context
- 262K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 18 · 7 core
Qwen3-Next 80B-A3B Thinking
Qwen · released 2025-09-11
text
Quality
Benchmark matrix
11 shared · 7 only Qwen3 4B Instruct 2507 · 5 only Qwen3-Next 80B-A3B Thinking| Benchmark | Qwen3 4B Instruct 2507 | Qwen3-Next 80B-A3B Thinking | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 11 | ||||
| MMLU-Pro | 69.6% ↗ | 82.7% ↗ | -13.1 pt | Qwen3-Next 80B-A3B Thinking |
| AIME25 | 47.4% ↗ | 87.8% ↗ | -40.4 pt | Qwen3-Next 80B-A3B Thinking |
| Arena-Hard v2 | 43.4% ↗ | 62.3% ↗ | -18.9 pt | Qwen3-Next 80B-A3B Thinking |
| BFCL-v3 | 61.9% ↗ | 72% ↗ | -10.1 pt | Qwen3-Next 80B-A3B Thinking |
| GPQA | 62% ↗ | 77.2% ↗ | -15.2 pt | Qwen3-Next 80B-A3B Thinking |
| HMMT25 | 31% ↗ | 73.9% ↗ | -42.9 pt | Qwen3-Next 80B-A3B Thinking |
| IFEval | 83.4% ↗ | 88.9% ↗ | -5.5 pt | Qwen3-Next 80B-A3B Thinking |
| LiveCodeBench v6 (25.02-25.05) | 35.1% ↗ | 68.7% ↗ | -33.6 pt | Qwen3-Next 80B-A3B Thinking |
| MMLU-Redux | 84.2% ↗ | 92.5% ↗ | -8.3 pt | Qwen3-Next 80B-A3B Thinking |
| SuperGPQA | 42.8% ↗ | 60.8% ↗ | -18 pt | Qwen3-Next 80B-A3B Thinking |
| TAU2-Retail | 40.4% ↗ | 67.8% ↗ | -27.4 pt | Qwen3-Next 80B-A3B Thinking |
| Only Qwen3 4B Instruct 2507 reports · 7 | ||||
| GPQA Diamond | 45.8% ↗epoch run | not reported | — | — |
| Aider-Polyglot | 12.9% ↗ | not reported | — | — |
| Chess Puzzles (Epoch AI run) | 4% ↗epoch run | not reported | — | — |
| LiveBench 20241125 | 63% ↗ | not reported | — | — |
| MultiPL-E | 76.8% ↗ | not reported | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 52.2% ↗epoch run | not reported | — | — |
| ZebraLogic | 80.2% ↗ | not reported | — | — |
| Only Qwen3-Next 80B-A3B Thinking reports · 5 | ||||
| LMArena Elo | not reported | 1369.4 ↗ | — | — |
| LiveBench 241125 | not reported | 76.6% ↗ | — | — |
| OJBench | not reported | 29.7% ↗ | — | — |
| TAU2-Airline | not reported | 60.5% ↗ | — | — |
| TAU2-Telecom | not reported | 43.9% ↗ | — | — |
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 Qwen3 4B Instruct 2507 minus Qwen3-Next 80B-A3B Thinking 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 | Qwen3 4B Instruct 2507 | Qwen3-Next 80B-A3B Thinking | Edge |
|---|---|---|---|
| Context window | 262K tokens | 262K tokens | Tie |
| Max output | — | — | — |
| Input price / 1M | — | $0.15 ↗ | — |
| Output price / 1M | — | $1.2 ↗ | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | $1.35 | — |
| Modalities | text | text | Tie |
| Released | 2025-08-06 | 2025-09-11 | — |
| Cited benchmark scores | 18 | 16 | — |
Reliability
Qwen status
Live from /statusMore matchups
Qwen3 4B Instruct 2507 vs …
Models sharing the most benchmarksMore matchups
Qwen3-Next 80B-A3B Thinking vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Qwen3 4B Instruct 2507 and Qwen3-Next 80B-A3B Thinking on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.