GLM-5.3-Flash vs GPT-5.6 Luna
GPT-5.6 Luna leads GLM-5.3-Flash by 3.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GLM-5.3-Flash wins 5 and GPT-5.6 Luna wins 10 of 15 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | GLM-5.3-Flash | GPT-5.6 Luna |
|---|---|---|
| Agentic coding | 54 | 59 |
| Long-horizon autonomy | 57 | 61 |
| Knowledge-work agents | 40 | 39 |
| Computer use | 34 | 40 |
| Security & context learning | 47 | 51 |
Shared benchmarks
| Benchmark | GLM-5.3-Flash | GPT-5.6 Luna |
|---|---|---|
| APEX-Agentsindex | 52.8% | 43.0% |
| DeepSWEindex | 63.4% | 67.2% |
| CursorBenchindex | 36.8% | 35.9% |
| ARC-AGI | 91.0% | 88.0% |
| ARC-AGI-2 | 65.8% | 59.5% |
| Chess Puzzles | 21.0% | 43.0% |
| CritPt | 15.4% | 20.6% |
| FrontierCode | 31.8% | 39.8% |
| FrontierMath Tier 4 | 17.1% | 61.0% |
| FrontierMath Tiers 1-3 | 55.8% | 82.1% |
| GPQA Diamond | 90.2% | 91.6% |
| OTIS Mock AIME | 93.9% | 98.3% |
| ProofBench | 21.0% | 60.0% |
| SciCode | 51.6% | 53.6% |
| WebDev Arena | 1607 | 1519 |
Specs and price
| GLM-5.3-Flash | GPT-5.6 Luna | |
|---|---|---|
| Agent Index | 48.3 (#38) | 51.4 (#34) |
| Agent $/M | $0.16 | $0.10 |
| Input / output $/M | $0.30 / $1.05 | $0.20 / $1.20 |
| Context | 1M | 1.1M |
| Released | Sep 18, 2026 | Jul 9, 2026 |
| Weights | Open | Closed |
| API hosts | 42 | 5 |
Don’t pick one. Route.
Shogo can run GLM-5.3-Flash and GPT-5.6 Luna side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo