Model comparison
Gemini 2.5 Flash vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 51.1 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 2.5× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
GoogleSummary
- They share 30 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 1 category and GLM-5.2 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 18.1.
- The biggest single-benchmark swing is ARC-AGI-1: 33.3% for Gemini 2.5 Flash and 77% for GLM-5.2.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1M.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | GLM-5.2 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | |
| Noometry Index | 39.3 | 51.1 |
| Released | 2025-04-17 | 2026-06-13 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.30 | $1.40 |
| Output $ / M tokens | $2.50 | $4.40 |
| Results tracked | 54 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
Gemini 2.5 Flash: 35.8 (#220), GLM-5.2: 51.3 (#41)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| WeirdML | 41.9% | 70.1% |
| LMArena Coding | 1424 | 1485 |
| ALE-Bench | 661.88 | 1,047 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1603 |
| SciCode | — | 50.5% |
Agentic & Tool Use GLM-5.2 leads
Gemini 2.5 Flash: 30.8 (#74), GLM-5.2: 32.4 (#63)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| Vending-Bench 2 | 548.84 | 8,314 |
| Terminal-Bench | 17.1% | — |
| APEX-Agents | — | 45.2% |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 31.7% |
| BALROG | 33.5% | — |
| GBAEval | — | 0% |
Reasoning GLM-5.2 leads
Gemini 2.5 Flash: 18.1 (#286), GLM-5.2: 42.3 (#52)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 2.5% | 22.8% |
| SimpleBench | 41.2% | 58.8% |
| Kagi LLM Benchmark | 56.8% | 62.6% |
| ARC-AGI-1 | 33.3% | 77% |
| CritPt | 1.1% | 20.9% |
| LMArena Hard Prompts | 1422 | 1480 |
| DTBench | 76.5% | 93.6% |
| LMCA | 27.5% | 45.8% |
| Epoch Capabilities Index | 143.03 | 151.78 |
| NYT Connections (extended) | — | 74.3% |
| Chess Puzzles | — | 21% |
| EnigmaEval | 2.7% | — |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 60.6 | — |
Math GLM-5.2 leads
Gemini 2.5 Flash: 39.9 (#98), GLM-5.2: 55.7 (#43)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 86.4% |
| LMArena Math | 1415 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| Omni-MATH | 38.5% | — |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge GLM-5.2 leads
Gemini 2.5 Flash: 36.4 (#168), GLM-5.2: 57.1 (#40)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Expert | 1426 | 1486 |
| GPQA Diamond | — | 91.9% |
| Humanity's Last Exam | 12.1% | — |
| SimpleQA Verified | — | 34.2% |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
Multimodal Not comparable
Gemini 2.5 Flash: 41.8 (#32), GLM-5.2: —
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1253 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual GLM-5.2 leads
Gemini 2.5 Flash: 52.3 (#88), GLM-5.2: 55.8 (#26)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1409 | 1459 |
| LMArena Chinese | 1450 | 1519 |
| LMArena French | 1433 | 1479 |
| LMArena German | 1418 | 1468 |
| LMArena Japanese | 1405 | 1451 |
| LMArena Korean | 1385 | 1445 |
| LMArena Russian | 1415 | 1466 |
| LMArena Spanish | 1421 | 1477 |
Instruction Following GLM-5.2 leads
Gemini 2.5 Flash: 75.7 (#54), GLM-5.2: 76.9 (#34)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1465 |
| IFEval | 89.8% | — |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), GLM-5.2: 45.3 (#43)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1419 | 1479 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference GLM-5.2 leads
Gemini 2.5 Flash: 53.8 (#157), GLM-5.2: 70.4 (#21)
| Benchmark | Gemini 2.5 Flash | GLM-5.2 |
|---|---|---|
| LMArena Text | 1417 | 1470 |
| LMArena Creative Writing | 1400 | 1462 |
| EQ-Bench Creative Writing | 1137 | 1757 |
| LMArena Multi-Turn | 1408 | 1469 |
| Short-Story Creative Writing | 76.5% | — |
| WildBench | 81.7% | — |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is Gemini 2.5 Flash better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 51.1 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 2.5× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash or GLM-5.2?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is Gemini 2.5 Flash or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 2.5 Flash and GLM-5.2 share?
30 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and GLM-5.2 has 51.