Model comparison
Gemini 2.5 Flash vs Qwen2.5 72B Instruct
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 31.9 on the Noometry Index.
Last verified . 31 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 31 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 8 categories and Qwen2.5 72B Instruct in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.5 Flash leads 39.9 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.1% for Gemini 2.5 Flash and 8.1% for Qwen2.5 72B Instruct.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 39.3 | 31.9 |
| Released | 2025-04-17 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 66K | 8K |
| Input $ / M tokens | $0.30 | $1.40 |
| Output $ / M tokens | $2.50 | $5.60 |
| Results tracked | 54 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Flash leads
Gemini 2.5 Flash: 35.8 (#220), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 41.9% | 16% |
| LMArena Coding | 1424 | 1292 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Gemini 2.5 Flash leads
Gemini 2.5 Flash: 30.8 (#74), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | 41.1% | 5.7% |
| BALROG | 33.5% | 16.2% |
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| METR Time Horizons | — | 35.8% |
| Vending-Bench 2 | 548.84 | — |
Reasoning Qwen2.5 72B Instruct leads
Gemini 2.5 Flash: 18.1 (#286), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1422 | 1271 |
| DTBench | 76.5% | 62.9% |
| LMCA | 27.5% | 13.4% |
| Epoch Capabilities Index | 143.03 | 129 |
| ForecastBench | 60.6 | 57.5 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| ARC-AGI-1 | 33.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 2.7% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Gemini 2.5 Flash leads
Gemini 2.5 Flash: 39.9 (#98), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 8.1% |
| Omni-MATH | 38.5% | 33% |
| LMArena Math | 1415 | 1283 |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 2.5 Flash leads
Gemini 2.5 Flash: 36.4 (#168), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| MMLU-Pro | 63.9% | 63.1% |
| Confabulations | 16.8% | 19.1% |
| GPQA (HELM) | 39% | 42.6% |
| LMArena Expert | 1426 | 1245 |
| GPQA Diamond | — | 49.1% |
| Humanity's Last Exam | 12.1% | — |
| Vectara Hallucination Rate | 7.8% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Gemini 2.5 Flash: 41.8 (#32), Qwen2.5 72B Instruct: —
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1253 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Gemini 2.5 Flash leads
Gemini 2.5 Flash: 52.3 (#88), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1409 | 1252 |
| LMArena Chinese | 1450 | 1272 |
| LMArena French | 1433 | 1280 |
| LMArena German | 1418 | 1234 |
| LMArena Japanese | 1405 | 1180 |
| LMArena Korean | 1385 | 1188 |
| LMArena Russian | 1415 | 1264 |
| LMArena Spanish | 1421 | 1256 |
Instruction Following Gemini 2.5 Flash leads
Gemini 2.5 Flash: 75.7 (#54), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 89.8% | 80.6% |
| LMArena Instruction Following | 1405 | 1254 |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1419 | 1282 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference Gemini 2.5 Flash leads
Gemini 2.5 Flash: 53.8 (#157), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Gemini 2.5 Flash | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1417 | 1269 |
| LMArena Creative Writing | 1400 | 1221 |
| WildBench | 81.7% | 80.2% |
| LMArena Multi-Turn | 1408 | 1272 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Qwen2.5 72B Instruct?
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 31.9 on the Noometry Index.
Which is cheaper, Gemini 2.5 Flash or Qwen2.5 72B Instruct?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Gemini 2.5 Flash or Qwen2.5 72B Instruct better for coding?
Gemini 2.5 Flash scores higher on coding benchmarks: 35.8 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 131K.
How many benchmarks do Gemini 2.5 Flash and Qwen2.5 72B Instruct share?
31 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Qwen2.5 72B Instruct has 43.