Model comparison
Qwen2.5 7B Instruct vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 4.0× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.
Last verified . 11 shared benchmarks.
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. Qwen2.5 7B Instruct scores higher in 0 categories and Qwen3 235B-A22B in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 50.4 to 12.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Qwen2.5 7B Instruct and 86.7% for Qwen3 235B-A22B.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
Side by side
| Qwen2.5 7B Instruct | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 29.0 | 43.5 |
| Released | 2024-09 | 2025-04 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.17 | $0.70 |
| Output $ / M tokens | $0.70 | $2.80 |
| Results tracked | 15 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 36.5 (#208), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1445 |
| BigCodeBench Complete | 46.1% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 23.8 (#124), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| BALROG | 7.8% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Too close to call
Qwen2.5 7B Instruct: 14.8 (#322), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| Chess Puzzles | 0% | 12% |
| DTBench | 47.7% | 80.3% |
| LMCA | 6.4% | 29.3% |
| Epoch Capabilities Index | 118.51 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| LMArena Hard Prompts | — | 1433 |
| Mystery Game Puzzles | — | 9% |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 12.6 (#306), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 86.7% |
| Omni-MATH | 29.4% | 71.8% |
| LMArena Math | — | 1432 |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 17.0 (#286), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 35.5% | 80.1% |
| MMLU-Pro | 53.9% | 84.4% |
| GPQA (HELM) | 34.1% | 72.7% |
| SimpleQA Verified | — | 40.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| LMArena Expert | — | 1463 |
| MMLU | 72.9% | — |
Multilingual Not comparable
Qwen2.5 7B Instruct: —, Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | — | 1409 |
| LMArena Chinese | — | 1481 |
| LMArena French | — | 1445 |
| LMArena German | — | 1433 |
| LMArena Japanese | — | 1399 |
| LMArena Korean | — | 1391 |
| LMArena Russian | — | 1411 |
| LMArena Spanish | — | 1430 |
Instruction Following Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 63.2 (#231), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 74.1% | 83.5% |
| LMArena Instruction Following | — | 1408 |
Long Context Not comparable
Qwen2.5 7B Instruct: —, Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| Fiction.LiveBench | — | 75% |
| LMArena Longer Query | — | 1426 |
Writing & Preference Qwen3 235B-A22B leads
Qwen2.5 7B Instruct: 48.8 (#195), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Qwen2.5 7B Instruct | Qwen3 235B-A22B |
|---|---|---|
| WildBench | 73.1% | 86.6% |
| LMArena Text | — | 1419 |
| LMArena Creative Writing | — | 1384 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| LMArena Multi-Turn | — | 1432 |
Frequently asked questions
Is Qwen2.5 7B Instruct better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 4.0× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.
Which is cheaper, Qwen2.5 7B Instruct or Qwen3 235B-A22B?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is Qwen2.5 7B Instruct or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 36.5 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Qwen2.5 7B Instruct and Qwen3 235B-A22B share?
11 benchmarks have published results for both models. Qwen2.5 7B Instruct has 15 scored results on Noometry and Qwen3 235B-A22B has 49.