Mercury 3
| Released | 2026-10-16 |
|---|---|
| Price per 1M | $0.10 in · $0.35 out |
| Overall score | 56.6 · rank 60 |
| Context window | 1,000,000 tokens |
| Provider | Inception |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 56.6 | 60 | 8 |
| Overall: Coding | 57.1 | 57 | 7 |
| Overall: Reasoning | 55.0 | 74 | 1 |
| Overall: Agentic work | 56.6 | 63 | 6 |
| Overall: Frontend | 55.7 | 57 | 2 |
| Coding | 58.3 | 44 | 1 |
| Agents | 55.9 | 69 | 5 |
| Knowledge | 55.0 | 59 | 1 |
| Frontend | 54.4 | 57 | 1 |
| Value vs price | +14.3 | 10 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | – | |
| High | 56.6 ±2.9 | 8 |
Evidence
Coding 58.3 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Code Migration v1.1 | 39.3% | Vals AI | 2026-11-09 |
Agents 55.9 ±1.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4.0 (Vals run) | 3.0% | Vals AI | 2026-11-09 |
| EMB v1.1 (Excel financial modeling) | 57.9% | Vals AI | 2026-11-09 |
| Tax Agent Bench | 18.2% | Vals AI | 2026-11-09 |
| Finance Agent v2.1 | 54.7% | Vals AI | 2026-11-09 |
| HLAB-Vals 1.1 (legal agent) | 5.0% | Vals AI | 2026-11-09 |
Knowledge 55.0 ±6.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Legal Research Bench | 28.8% | Vals AI | 2026-11-09 |
Frontend 54.4 ±0.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 60.9% | Vals AI | 2026-11-09 |
Names matched to this model
mercury-3