GPT-6 Luna
- Input
- $0.10/M tok
- Output
- $0.50/M tok
- Blended
- $0.20/M tok · 3:1
- Speed
- 132tok/s
- Context
- 1Mtokens
- Cost per task
- $0.15/task · AA
Writing: no LMArena/EQ-Bench entry or lab-reported writing benchmark as of 2026-09-24, so its Writing entry is fully estimated from its other categories (flagged “Estimated from other categories”).
Rankings
Coding
Verified#55 / 137rank by quality52.8±4.5quality / 1004/7 benchmarks measuredWriting & chat
Verified#44 / 162rank by quality64.0±6.0quality / 1002/5 benchmarks measuredVision
Verified#24 / 98rank by quality71.7±4.8quality / 1003/6 benchmarks measuredHard reasoning
Verified#48 / 218rank by quality67.8±5.1quality / 1002/4 benchmarks measuredAgents
Verified#38 / 187rank by quality63.7±7.8quality / 1002/7 benchmarks measured
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#3 of 21 (estimated)~82.7%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#9 of 24 (estimated)~63.5%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #8Hy4 preview65.7%
- ~#9GPT-6 LunaEstimated63.5%
- #9GPT-5.6 Terra63.4%
No published score — estimated from related benchmarks
DeepSWE v1.1Verified
#24 of 4563.7%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #22Hy4 preview64.3%
- #24GPT-6 Luna63.7%
- #25GLM-5.3 Flash63.4%
artificialanalysis.ai · 2026-09-24
Terminal-Bench 4.0Verified
#36 of 10112.6%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #36GPT-5.5 Instant (June 2026)12.6%
- #36GPT-6 Luna12.6%
- #36Kimi K312.6%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchEstimated
~#11 of 28 (estimated)~81.0%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #10Ternary Bonsai 27B82.8%
- ~#11GPT-6 LunaEstimated81.0%
- #11Gemma 4 31B80.0%
No published score — estimated from related benchmarks
SciCodeVerified
#29 of 11854.6%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #27GPT-5.6 Terra55.0%
- #29GPT-6 Luna54.6%
- #30Claude Sonnet 554.3%
artificialanalysis.ai · 2026-09-24
Writing & chat
IFBenchEstimated
~#42 of 140 (estimated)~68.7%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #41Mistral Medium 3.568.8%
- ~#42GPT-6 LunaEstimated68.7%
- #42Step 3.7 Flash67.3%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Verified
#27 of 561436 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #26Gemma 4 31BFew votes1437
- #27GPT-6 Luna1436
- #28Qwen3.5 397B A17B1434
lmarena.ai · 2026-09-25
EQ-Bench Creative Writing v3 (Elo)Estimated
~#10 of 26 (estimated)~1868 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #9Muse Spark 1.31906
- ~#10GPT-6 LunaEstimated1868
- #10GPT-5.6 Terra1855
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#17 of 8779.7%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #16Qwen3.8 Flash-Next79.8%
- #17GPT-6 Luna79.7%
- #18Apodex 1.179.2%
artificialanalysis.ai · 2026-10-02
CharXiv ReasoningEstimated
~#6 of 12 (estimated)~82.0%
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #5MiniMax-M3.1-Flash82.5%
- ~#6GPT-6 LunaEstimated82.0%
- #6Nemotron 3.5 Super VL80.8%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#4 of 8 (estimated)~88.1%
No published score — estimated from related benchmarks
GDP.pdf (AA)Verified
#11 of 4722.8%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #10GPT-5.6 Terra24.0%
- #11GPT-6 Luna22.8%
- #11Qwen3.8 Max (0902)22.8%
artificialanalysis.ai · 2026-10-10
Hard reasoning
GPQA DiamondEstimated
~#45 of 193 (estimated)~88.2%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #44DeepSeek-V4-Pro-0813-nvfp4-DSpark88.4%
- ~#45GPT-6 LunaEstimated88.2%
- #45Qwen3.8 27B NVFP488.0%
No published score — estimated from related benchmarks
Humanity's Last ExamVerified
#38 of 19838.5%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #37MiniMax-M339.0%
- #38GPT-6 Luna38.5%
- #39Grok Build 0.1 061638.3%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#10 of 34 (estimated)~91.8%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #9Qwen3.8 27B MXFP492.5%
- ~#10GPT-6 LunaEstimated91.8%
- #10GLM-5.3-MXFP491.3%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Estimated
~#24 of 120 (estimated)~86.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #23Qwen3.5 9B86.8%
- ~#24GPT-6 LunaEstimated86.8%
- #24GPT-5.6 Terra86.3%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#9 of 15 (estimated)~78.2%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #8GPT-5.578.7%
- ~#9GPT-6 LunaEstimated78.2%
- #9Gemini 3.5 Flash-Lite74.0%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~51.1%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#14 of 22 (estimated)~82.1%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #13Nex-N2.5-Mini83.4%
- ~#14GPT-6 LunaEstimated82.1%
- #14Nemotron 3.5 Super VL78.0%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#26 of 911432 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #25GPT-5.6 Terra1432
- #26GPT-6 Luna1432
- #27Mistral Large 41422
artificialanalysis.ai · 2026-10-10
AutomationBench-AAVerified
#25 of 8453.2%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #24Qwen3.8 Flash-Next55.9%
- #25GPT-6 Luna53.2%
- #26Step 5 Preview51.0%
artificialanalysis.ai · 2026-09-24
τ-Bench Banking (AA)Estimated
~#21 of 91 (estimated)~32.4%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #20K2 Horizon 375B A23B34.2%
- ~#21GPT-6 LunaEstimated32.4%
- #21K2 Horizon MoVA 36B A4B32.0%
No published score — estimated from related benchmarks
Not comparable across labs (1)
Kept for reference, not counted in rankings: these use a lab-specific task set, answer key or scoring, so scores can't be compared fairly across labs. Only published scores are shown.