Model comparison
Gemini 3.8 Flash vs GLM-5.3-Flash
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 51.8 on the Noometry Index. GLM-5.3-Flash costs 6.3× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Last verified . 39 shared benchmarks.
GoogleSummary
- They share 39 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and GLM-5.3-Flash in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 48.0.
- The biggest single-benchmark swing is Chess Puzzles: 61% for Gemini 3.8 Flash and 14% for GLM-5.3-Flash.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 1M.
- GLM-5.3-Flash has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.8 Flash | GLM-5.3-Flash | |
|---|---|---|
| Provider | Z.ai (Zhipu) | |
| Noometry Index | 61.8 | 51.8 |
| Released | 2026-09-02 | 2026-08-20 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.75 | $0.15 |
| Output $ / M tokens | $3.75 | $0.50 |
| Results tracked | 50 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), GLM-5.3-Flash: 53.1 (#31)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| DeepSWE | 73.8% | 63.4% |
| FrontierCode | 41.2% | 31.8% |
| CursorBench | 39.6% | 36.8% |
| LMArena WebDev | 1584 | 1609 |
| FrontierSWE | 19.6% | 18.1% |
| SciCode | 56.6% | 51.6% |
| LMArena Coding | 1510 | 1508 |
| ALE-Bench | 1,270 | 303.55 |
| WeirdML | 84.8% | — |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.8 Flash: 41.8 (#21), GLM-5.3-Flash: 34.2 (#47)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| APEX-Agents | 64.3% | 52.8% |
| GDP.pdf | 23.4% | 14% |
| Remote Labor Index | 5.8% | — |
| Vending-Bench 2 | 5,094 | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), GLM-5.3-Flash: 48.0 (#42)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| ARC-AGI-2 | 89.2% | 65.8% |
| ARC-AGI-1 | 98.5% | 91% |
| CritPt | 18.3% | 15.4% |
| Chess Puzzles | 61% | 14% |
| LMArena Hard Prompts | 1508 | 1491 |
| Mystery Game Puzzles | 47% | 8% |
| Surface Evolver Bench | 76.9% | 52.5% |
| Epoch Capabilities Index | 156.71 | 151.88 |
| NYT Connections (extended) | 97.4% | — |
| DTBench | 95.7% | — |
| LMCA | 52.9% | — |
| Bench to the Future 3 | — | 0.15 |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), GLM-5.3-Flash: 53.3 (#47)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 55.8% |
| FrontierMath Tier 4 | 22% | 17.1% |
| OTIS Mock AIME 2024-2025 | 98.9% | 93.9% |
| ProofBench | 48% | 21% |
| LMArena Math | 1528 | 1500 |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), GLM-5.3-Flash: 58.4 (#36)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| GPQA Diamond | 95.4% | 90.2% |
| LMArena Expert | 1524 | 1513 |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
Multimodal GLM-5.3-Flash leads
Gemini 3.8 Flash: 40.7 (#45), GLM-5.3-Flash: 42.8 (#27)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| LMArena Vision | 1314 | 1296 |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), GLM-5.3-Flash: 56.0 (#25)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| LMArena Non-English | 1491 | 1462 |
| LMArena Chinese | 1554 | 1527 |
| LMArena French | 1498 | 1496 |
| LMArena German | 1493 | 1470 |
| LMArena Japanese | 1502 | 1429 |
| LMArena Korean | 1459 | 1446 |
| LMArena Russian | 1515 | 1469 |
| LMArena Spanish | 1485 | 1471 |
Instruction Following Too close to call
Gemini 3.8 Flash: 78.0 (#13), GLM-5.3-Flash: 77.5 (#20)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| LMArena Instruction Following | 1490 | 1478 |
Long Context Too close to call
Gemini 3.8 Flash: 46.3 (#24), GLM-5.3-Flash: 45.4 (#39)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| LMArena Longer Query | 1508 | 1482 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), GLM-5.3-Flash: 65.3 (#50)
| Benchmark | Gemini 3.8 Flash | GLM-5.3-Flash |
|---|---|---|
| LMArena Text | 1499 | 1471 |
| LMArena Creative Writing | 1492 | 1442 |
| LMArena Multi-Turn | 1501 | 1467 |
| EQ-Bench Creative Writing | 1748 | — |
Frequently asked questions
Is Gemini 3.8 Flash better than GLM-5.3-Flash?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 51.8 on the Noometry Index. GLM-5.3-Flash costs 6.3× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.8 Flash or GLM-5.3-Flash?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.
Is Gemini 3.8 Flash or GLM-5.3-Flash better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 53.1 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 3.8 Flash and GLM-5.3-Flash share?
39 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and GLM-5.3-Flash has 40.