GPT-6 LunavsGLM-5.3-Flash
GPT-6 Luna and GLM-5.3-Flash are tied across 4 shared benchmarks: GPT-6 Luna leads on 2, GLM-5.3-Flash leads on 2, with 0 ties and an average score difference of -72.07.
GPT-6 Luna
OpenAI · 2026-09-22 · Reasoning model
GLM-5.3-Flash
智谱AI · 2026-08-26 · Multimodal model
GPT-6 Luna2 wins(50%)(50%)2 winsGLM-5.3-Flash
Benchmark scores
Grouped by capability, sorted by largest gap within each. 4 shared benchmarks.
Productivity Knowledge
GLM-5.3-Flash 2/2| Benchmark | GPT-6 Luna | GLM-5.3-Flash | Diff |
|---|---|---|---|
| GDPval-AA v2 | 1,36758 / 110Max (With Tools) | 1,65512 / 110Max (With Tools) | -288 |
| AutomationBench | 20.7022 / 23Max (With Tools) | 48.807 / 23Max (With Tools) | -28.10 |
Agent Level Benchmark
GPT-6 Luna 1/1| Benchmark | GPT-6 Luna | GLM-5.3-Flash | Diff |
|---|---|---|---|
| Agents' Last Exam | 50.905 / 24Max (With Tools) | 26.3018 / 24Max (With Tools) | +24.60 |
Coding and Software Engineer
GPT-6 Luna 1/1| Benchmark | GPT-6 Luna | GLM-5.3-Flash | Diff |
|---|---|---|---|
| DeepSWE | 66.6037 / 91Max (With Tools) | 63.3944 / 91Max (With Tools) | +3.21 |
Specs
| Field | GPT-6 Luna | GLM-5.3-Flash |
|---|---|---|
| Publisher | OpenAI | 智谱AI |
| Release date | 2026-09-22 | 2026-08-26 |
| Model type | Reasoning model | Multimodal model |
| Architecture | Dense | MoE |
| Parameters | Not available | 320B |
| Context length | 1.05M | 1M |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-6 Luna | GLM-5.3-Flash |
|---|---|---|
| Text input | $0.1 / 1M tokens | $0.075 / 1M tokens |
| Text output | $0.5 / 1M tokens | $0.25 / 1M tokens |
| Cache read | $0.01 / 1M tokens | $0.015 / 1M tokens |
| Cache write | $0.125 / 1M tokens | Not public |
Summary
- GPT-6 Lunaleads in:Agent Level Benchmark (1/1), Coding and Software Engineer (1/1)
- GLM-5.3-Flashleads in:Productivity Knowledge (2/2)
On average across the 4 shared benchmarks, GLM-5.3-Flash scores 72.07 higher.
Largest single-benchmark gap: GDPval-AA v2 — GPT-6 Luna 1,367 vs GLM-5.3-Flash 1,655 (-288).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.