Model comparison
GPT-6 Luna vs Mercury 2.5
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.5 on the Noometry Index. Mercury 2.5 costs 3.0× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Summary
- They share 4 benchmarks with published results for both. GPT-6 Luna scores higher in 3 categories and Mercury 2.5 in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 23.3.
- The biggest single-benchmark swing is ProofBench: 64% for GPT-6 Luna and 3% for Mercury 2.5.
- Mercury 2.5 is cheaper at $0.04 / $0.15 per million input/output tokens, against $0.10 / $0.50 for GPT-6 Luna.
- GPT-6 Luna accepts more context: 1.05M tokens versus 260K.
Side by side
| GPT-6 Luna | Mercury 2.5 | |
|---|---|---|
| Provider | OpenAI | Inception |
| Noometry Index | 53.3 | 33.5 |
| Released | 2026-09-22 | 2026-09-08 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 260K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.10 | $0.04 |
| Output $ / M tokens | $0.50 | $0.15 |
| Results tracked | 42 | 4 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Mercury 2.5: 39.5 (#156)
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| SciCode | 54.6% | 38.5% |
| ALE-Bench | 1,577 | 301.65 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1581 | — |
| LMArena Coding | 1439 | — |
Agentic & Tool Use Not comparable
GPT-6 Luna: 33.3 (#54), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Mercury 2.5: 22.4 (#193)
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| CritPt | 19.4% | 0% |
| ARC-AGI-2 | 59.3% | — |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| Chess Puzzles | 31% | — |
| LMArena Hard Prompts | 1411 | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LMCA | 44.5% | — |
| Epoch Capabilities Index | 156.28 | — |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Mercury 2.5: 23.3 (#272)
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| ProofBench | 64% | 3% |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| LMArena Math | 1416 | — |
Knowledge Not comparable
GPT-6 Luna: 57.0 (#41), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 41.4% | — |
| LMArena Expert | 1444 | — |
Multimodal Not comparable
GPT-6 Luna: 42.4 (#30), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
Multilingual Not comparable
GPT-6 Luna: 50.5 (#117), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| LMArena Non-English | 1386 | — |
| LMArena Chinese | 1433 | — |
| LMArena French | 1420 | — |
| LMArena German | 1369 | — |
| LMArena Japanese | 1369 | — |
| LMArena Korean | 1360 | — |
| LMArena Russian | 1394 | — |
| LMArena Spanish | 1393 | — |
Instruction Following Not comparable
GPT-6 Luna: 74.3 (#99), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| LMArena Instruction Following | 1409 | — |
Long Context Not comparable
GPT-6 Luna: 43.0 (#111), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| LMArena Longer Query | 1409 | — |
Writing & Preference Not comparable
GPT-6 Luna: 58.3 (#119), Mercury 2.5: —
| Benchmark | GPT-6 Luna | Mercury 2.5 |
|---|---|---|
| LMArena Text | 1391 | — |
| LMArena Creative Writing | 1363 | — |
| LMArena Multi-Turn | 1396 | — |
Frequently asked questions
Is GPT-6 Luna better than Mercury 2.5?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.5 on the Noometry Index. Mercury 2.5 costs 3.0× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Which is cheaper, GPT-6 Luna or Mercury 2.5?
Mercury 2.5 is cheaper. It lists at $0.04 per million input tokens and $0.15 per million output tokens; GPT-6 Luna lists at $0.10 and $0.50.
Is GPT-6 Luna or Mercury 2.5 better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 39.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 260K.
How many benchmarks do GPT-6 Luna and Mercury 2.5 share?
4 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Mercury 2.5 has 4.