ExamenosMage-VL-Cowork
Explore Mage-VL-Cowork from Microsoft: published specifications, source-linked vendor benchmarks, independent evaluator coverage and recorded pricing when available.
Compare Mage-VL-Cowork with other models →Compare subscription plansExplore data coverage
Pricing · OpenRouter
No OpenRouter pricing is recorded for this model. Missing rates are not free.
Official / vendor benchmarks
Default headline records. Own-vendor, peer-vendor and third-party provenance remain visible in evidence; configurations may differ.
| Benchmark / evaluator | Headline score | Evidence |
|---|---|---|
| OSWorld-Verified (Mage 361, 50 steps) | 66%unknown effort | All 1 recorded result & sources66% · raw 66.04 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · OS | 87.5%unknown effort | All 1 recorded result & sources87.5% · raw 87.5 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. Domain value as printed in the card table. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Office | 64.9%unknown effort | All 1 recorded result & sources64.9% · raw 64.9 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. Domain value as printed in the card table. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Daily | 65%unknown effort | All 1 recorded result & sources65% · raw 65 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. Domain value as printed in the card table. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Professional | 66.3%unknown effort | All 1 recorded result & sources66.3% · raw 66.3 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. Domain value as printed in the card table. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Workflow | 62.7%unknown effort | All 1 recorded result & sources62.7% · raw 62.7 % Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. Domain value as printed in the card table. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Mean model calls per task | 18.8 calls/taskunknown effort | All 1 recorded result & sources18.8 calls/task · raw 18.77 calls/task Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Mean output tokens per task | 5499 tokens/taskunknown effort | All 1 recorded result & sources5499 tokens/task · raw 5499 tokens/task Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. https://huggingface.co/microsoft/Mage-VL-Cowork |
| OSWorld-Verified (Mage 361, 50 steps) · Mean think tokens per task | 2482 tokens/taskunknown effort | All 1 recorded result & sources2482 tokens/task · raw 2482 tokens/task Headline · unknown effort · Own vendor Source/record date: 2026-10-30 Harness: MAGEHybridAgent, Microsoft-reported on the released Mage-VL-Cowork checkpoint (Stage 5 online hybrid RL), two-run average (65.33 and 66.75). Effort tiers not applicable; evaluation date unpublished. https://huggingface.co/microsoft/Mage-VL-Cowork |
Independent evaluators
Evaluator harnesses are distinct from vendor measurements. Missing coverage is not a failed test.
No independent evaluator results are recorded for this model.
Read how we select and source scores or the comparison guide.