Model comparison
Qwen2.5 7B Instruct vs Qwen3.7 Plus
Qwen3.7 Plus is the stronger model overall, scoring 45.0 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 2.3× less per token, which makes it the better buy when Qwen3.7 Plus's lead doesn't matter for your workload.
Last verified . 6 shared benchmarks.
Alibaba (Qwen)Summary
- They share 6 benchmarks with published results for both. Qwen2.5 7B Instruct scores higher in 1 category and Qwen3.7 Plus in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.7 Plus leads 54.9 to 17.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Qwen2.5 7B Instruct and 93.3% for Qwen3.7 Plus.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.40 / $1.60 for Qwen3.7 Plus.
- Qwen3.7 Plus accepts more context: 1M tokens versus 131K.
- Qwen2.5 7B Instruct has downloadable open weights; the other is API-only.
Side by side
| Qwen2.5 7B Instruct | Qwen3.7 Plus | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 28.6 | 45.0 |
| Released | 2024-09 | 2026-06-02 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.17 | $0.40 |
| Output $ / M tokens | $0.70 | $1.60 |
| Results tracked | 15 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Qwen2.5 7B Instruct: 35.9 (#204), Qwen3.7 Plus: 36.1 (#200)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| FrontierCode | — | 10.2% |
| SciCode | — | 45.5% |
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1473 |
| BigCodeBench Complete | 46.1% | — |
Agentic & Tool Use Qwen2.5 7B Instruct leads
Qwen2.5 7B Instruct: 23.5 (#117), Qwen3.7 Plus: 21.2 (#130)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| OSWorld 2.0 | — | 2.8% |
| BALROG | 7.8% | — |
Reasoning Qwen3.7 Plus leads
Qwen2.5 7B Instruct: 14.3 (#317), Qwen3.7 Plus: 38.4 (#54)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| Chess Puzzles | 0% | 24% |
| DTBench | 47.7% | 84% |
| LMCA | 6.4% | 37.6% |
| Epoch Capabilities Index | 118.44 | 147.39 |
| NYT Connections (extended) | — | 74.8% |
| CritPt | — | 9.1% |
| LMArena Hard Prompts | — | 1460 |
| Mystery Game Puzzles | — | 17% |
Math Qwen3.7 Plus leads
Qwen2.5 7B Instruct: 11.8 (#300), Qwen3.7 Plus: 49.3 (#47)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 93.3% |
| FrontierMath (Tiers 1-3) | — | 34.4% |
| Omni-MATH | 29.4% | — |
| LMArena Math | — | 1467 |
Knowledge Qwen3.7 Plus leads
Qwen2.5 7B Instruct: 17.0 (#281), Qwen3.7 Plus: 54.9 (#46)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| GPQA Diamond | 35.5% | 87.9% |
| MMLU-Pro | 53.9% | — |
| GPQA (HELM) | 34.1% | — |
| LMArena Expert | — | 1467 |
| MMLU | 72.9% | — |
Multimodal Not comparable
Qwen2.5 7B Instruct: —, Qwen3.7 Plus: 41.2 (#30)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| LMArena Vision | — | 1281 |
| LMArena Document | — | 1444 |
Multilingual Not comparable
Qwen2.5 7B Instruct: —, Qwen3.7 Plus: 55.0 (#36)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| LMArena Non-English | — | 1445 |
| LMArena Chinese | — | 1510 |
| LMArena French | — | 1471 |
| LMArena German | — | 1471 |
| LMArena Japanese | — | 1411 |
| LMArena Korean | — | 1416 |
| LMArena Russian | — | 1457 |
| LMArena Spanish | — | 1453 |
Instruction Following Qwen3.7 Plus leads
Qwen2.5 7B Instruct: 63.4 (#226), Qwen3.7 Plus: 75.9 (#49)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| IFEval | 74.1% | — |
| LMArena Instruction Following | — | 1439 |
Long Context Not comparable
Qwen2.5 7B Instruct: —, Qwen3.7 Plus: 44.5 (#61)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| LMArena Longer Query | — | 1454 |
Writing & Preference Qwen3.7 Plus leads
Qwen2.5 7B Instruct: 48.8 (#194), Qwen3.7 Plus: 64.5 (#54)
| Benchmark | Qwen2.5 7B Instruct | Qwen3.7 Plus |
|---|---|---|
| LMArena Text | — | 1454 |
| LMArena Creative Writing | — | 1439 |
| WildBench | 73.1% | — |
| LMArena Multi-Turn | — | 1459 |
Frequently asked questions
Is Qwen2.5 7B Instruct better than Qwen3.7 Plus?
Qwen3.7 Plus is the stronger model overall, scoring 45.0 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 2.3× less per token, which makes it the better buy when Qwen3.7 Plus's lead doesn't matter for your workload.
Which is cheaper, Qwen2.5 7B Instruct or Qwen3.7 Plus?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Qwen3.7 Plus lists at $0.40 and $1.60.
Is Qwen2.5 7B Instruct or Qwen3.7 Plus better for coding?
They score almost the same on coding (35.9 vs 36.1); test both on your own repository before choosing.
Which has the bigger context window?
Qwen3.7 Plus does, with 1M tokens against 131K.
How many benchmarks do Qwen2.5 7B Instruct and Qwen3.7 Plus share?
6 benchmarks have published results for both models. Qwen2.5 7B Instruct has 15 scored results on Noometry and Qwen3.7 Plus has 32.