GPT-6.1 Terra vs Muse Wonder
GPT-6.1 Terra leads Muse Wonder by 8.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-6.1 Terra wins 6 and Muse Wonder wins 3 of 9 shared benchmarks. GPT-6.1 Terra is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | GPT-6.1 Terra | Muse Wonder |
|---|---|---|
| Agentic coding | 79 | 68 |
| Long-horizon autonomy | 74 | 70 |
| Knowledge-work agents | 56 | 50 |
| Computer use | 63 | 50 |
| Security & context learning | 62 | 57 |
Shared benchmarks
| Benchmark | GPT-6.1 Terra | Muse Wonder |
|---|---|---|
| APEX-Agentsindex | 68.3% | 66.1% |
| Vending-Bench 2index | $9,107 | $9,432 |
| CritPt | 28.0% | 28.3% |
| FrontierMath Tier 4 | 92.7% | 70.7% |
| FrontierMath Tiers 1-3 | 90.9% | 82.1% |
| MirrorCode | 76.7% | 45.6% |
| OTIS Mock AIME | 100.0% | 99.4% |
| SciCode | 55.1% | 60.3% |
| WeirdML v3 | 28.1% | 15.1% |
Specs and price
| GPT-6.1 Terra | Muse Wonder | |
|---|---|---|
| Agent Index | 69 (#8) | 60.9 (#15) |
| Agent $/M | $0.42 | $1.69 |
| Input / output $/M | $1.00 / $5.00 | $4.00 / $20.00 |
| Context | 1.1M | 1M |
| Released | Oct 26, 2026 | Oct 12, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 2 |
Don’t pick one. Route.
Shogo can run GPT-6.1 Terra and Muse Wonder side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo