Native multimodal GLM model for efficient coding and long-horizon agent tasks
Benchmarks
| Benchmark | Category | Score | Variant | Provenance | Source | Observed |
|---|---|---|---|---|---|---|
| AA Agentic Index | agentic | 50.9 | — | mirror | openrouter | Sep 20, 2026 |
| 50.9 | — | independent | aa | Sep 20, 2026 | ||
| AA Coding Index | coding | 71.5 | — | mirror | openrouter | Aug 27, 2026 |
| 71.5 | — | independent | aa | Aug 27, 2026 | ||
| AA Intelligence Index | intelligence | 27.8 | — | mirror | openrouter | Oct 30, 2026 |
| 27.8 | — | independent | aa | Oct 30, 2026 | ||
| Design Arena — 3d | preference | 1318 | — | mirror | openrouter | Oct 16, 2026 |
| 1286 | — | mirror | openrouter | Nov 19, 2026 | ||
| Design Arena — asciiart | preference | 1264 | — | mirror | openrouter | Oct 16, 2026 |
| 1265 | — | mirror | openrouter | Oct 17, 2026 | ||
| 1255 | — | mirror | openrouter | Nov 18, 2026 | ||
| Design Arena — codecategories | preference | 1284 | — | mirror | openrouter | Oct 16, 2026 |
| 1262 | — | mirror | openrouter | Nov 19, 2026 | ||
| Design Arena — dataviz | preference | 1255 | — | mirror | openrouter | Nov 18, 2026 |
| Design Arena — gamedev | preference | 1283 | — | mirror | openrouter | Oct 16, 2026 |
| 1282 | — | mirror | openrouter | Oct 17, 2026 | ||
| 1260 | — | mirror | openrouter | Nov 19, 2026 | ||
| Design Arena — svg | preference | 1266 | — | mirror | openrouter | Nov 19, 2026 |
| Design Arena — uicomponent | preference | 1319 | — | mirror | openrouter | Oct 17, 2026 |
| 1292 | — | mirror | openrouter | Nov 19, 2026 | ||
| Design Arena — website | preference | 1273 | — | mirror | openrouter | Oct 16, 2026 |
| 1253 | — | mirror | openrouter | Nov 18, 2026 | ||
| LMArena Text | preference | 1471 | — | crowd | lmarena | Nov 9, 2026 |
| LMArena Vision | preference | 1295 | — | crowd | lmarena | Nov 9, 2026 |
| LMArena WebDev | preference | 1609 | — | crowd | lmarena | Nov 11, 2026 |
| Median output speed | speed | 55 tok/s | — | independent | aa | Nov 14, 2026 |
| Median time to first token | speed | 2.28s | — | independent | aa | Nov 14, 2026 |
Includes index data from Artificial Analysis.
Providers
Live operational stats per hosting endpoint, via OpenRouter.
| Provider | Quant | Context | $/1M in · out | Uptime 24h | Latency | Throughput |
|---|---|---|---|---|---|---|
| Baidu | fp8 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| BaseTen | fp8 | 1.0M | $0.15 · $0.5 | 99.86% | — | — |
| Cohere | nvfp4 | 1.0M | $0.15 · $0.5 | 99.65% | — | — |
| CoreWeave | nvfp4 | 1.0M | $0.12 · $0.38 | 99.03% | — | — |
| Crusoe | fp4 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Darkbloom | fp4 | 262K | $0.065 · $0.18 | 95.05% | — | — |
| DeepInfra | fp4 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| DekaLLM | — | 1.0M | $0.1 · $0.5 | 99.06% | — | — |
| DigitalOcean | — | 1.0M | $0.112 · $0.375 | 100.00% | — | — |
| Fireworks | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Fireworks | — | 1.0M | $0.225 · $0.75 | 100.00% | — | — |
| Friendli | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| GMICloud | fp8 | 1.0M | $0.075 · $0.25 | 99.86% | — | — |
| Inceptron | fp8 | 1.0M | $0.16 · $0.49 | 100.00% | — | — |
| InferenceNet | fp4 | 1.0M | $0.108 · $0.35 | 100.00% | — | — |
| InferenceNet | fp4 | 1.0M | $0.108 · $0.35 | 100.00% | — | — |
| KT | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Makora | fp4 | 1.0M | $0.14 · $0.24 | 100.00% | — | — |
| Mistral | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Modal | fp4 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Morph | fp8 | 1.0M | $0.106 · $0.307 | 99.03% | — | — |
| Near AI | fp8 | 1.0M | $0.08 · $0.5 | 99.93% | — | — |
| Nebius | fp4 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Novita | fp8 | 1.0M | $0.084 · $0.28 | 100.00% | — | — |
| OpenInference | fp4 | 1.0M | $0.06 · $0.75 | 100.00% | — | — |
| Parasail | fp4 | 1.0M | $0.188 · $0.625 | 100.00% | — | — |
| Parasail | fp8 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Particle AI | nvfp4 | 1.0M | $0.13 · $0.5 | 100.00% | — | — |
| Particle AI | nvfp4 | 1.0M | $0.065 · $0.25 | 100.00% | — | — |
| Reka | — | 1.0M | $0.12 · $0.3 | 100.00% | — | — |
| Relace | fp4 | 1.0M | $0.04 · $0.078 | 99.93% | — | — |
| River | — | 1.0M | $0.12 · $0.4 | 100.00% | — | — |
| Sail Research | fp4 | 1.0M | $0.175 · $0.625 | 100.00% | — | — |
| Sail Research | fp4 | 1.0M | $0.175 · $0.625 | 100.00% | — | — |
| SiliconFlow | fp8 | 1.0M | $0.1 · $0.4 | 92.24% | — | — |
| StreamLake | fp8 | 1.0M | $0.089 · $0.295 | 98.82% | — | — |
| Tenstorrent | fp8 | 262K | $0.075 · $0.25 | 99.27% | — | — |
| Together | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Venice | — | 1.0M | $0.15 · $0.5 | 100.00% | — | — |
| Wafer | nvfp4 | 1.0M | $0.065 · $0.24 | 100.00% | — | — |
| Wafer | nvfp4 | 1.0M | $0.042 · $0.184 | 100.00% | — | — |
| Z.AI | fp8 | 1.0M | $0.15 · $0.5 | 100.00% | — | — |