GPT-6 Luna

+ contribute

GPT-6 Luna is a closed-weights language model from OpenAI, released September 2026.

76.8±0.7#28 of 719 LLMsfrom 18 benchmark results across 6 sources

$ how it compares

The LLMs ranked either side of GPT-6 Luna, out of 828 with a score, then more from OpenAI.

#	model	score	where	RAM	released
26	GPT-5.2 Pro OpenAI	77±1	api $57.75/M		2025-12
27	MiMo-V2.6-Flash (309B-A15B) Xiaomi	77±1	api $0.17/M +local	~136GB	2026-09
28	GPT-6 Luna OpenAI	77±1	api $0.20/M		2026-09
29	GPT-5.6 Luna OpenAI	77±1	api $0.45/M		2026-07
30	Grok 4.6 SpaceXAI (xAI)	77±1	api $3.00/M		2026-08
	more from OpenAI
1	GPT-6 Astra OpenAI	87±1	api $20.00/M		2026-09
3	GPT-6.1 Sol OpenAI	85±1	api $4.00/M		2026-09
7	GPT-6 Sol OpenAI	83±1	api $4.00/M		2026-09

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #35 of 300 · 17% of its score · 3 tests

Trails GLM-5.3-Flash, Grok 4.7, Claude Opus 4.7 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arena Agent (October 2026 aggregate IPS)	0.0135 IPS ✎	▁▁▁▄▂▃▇█▂▂▁▃ beats 26/49	avg -0.312 IPS	▇▇▇▇ 9.0%	sharp
Terminal-Bench 4.0	16.4% ✎	█▅▅▅▅·▅▃▃▅▅▅ beats 4/20	very hard · avg 0.4%	▇▇ 5.2%	sharp
DeepSWE v1.1 (resolved) lab	66.6% ✎	█··▃▃····▃·▃ beats 0/6	hard · avg 21%	▇ 2.6%	at its level

coding▁▁▁▃▆██▇▆▄▃▁ #32 of 458 · 13% of its score · 3 tests

Beats GPT-5.5, GPT-5.6 Terra, Claude Opus 4.8 here, though they rank higher overall. Trails Qwen3.8 27B, GLM-5.2, GLM-5.3-Flash here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arena WebDev (Code Arena)	1578.7 ✎	▁▂▄▃▅█▄▇▆▄▁▁ beats 89/119	avg 981	▇▇▇ 8.0%	sharp
Luau Leetcode fully solved (Torpedo Software)	80.8% ✎	▁▁▁▁·▁▂▃▅▃█▇ beats 34/49	hard · avg 18%	▇ 2.8%	typical
DeepSWE v1.1 (resolved) lab	66.6% ✎	█··▃▃····▃·▃ beats 0/6	hard · avg 21%	▇ 2.6%	at its level

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #59 of 393 * · 4.5% of its score · 1 test

Trails Gemma 4 31B, Gemini 2.5 Pro, Claude Sonnet 4.5 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
LMArena (text)	1442.9 ✎	▁▁▁▂▃▃▃▆▆▇█▃ beats 304/379	avg 1268	▇▇ 4.5%	blunt

math▁▂▃▅▆▅▄█▄▃▁▁ #16 of 398 · 40% of its score · 3 tests

Beats Qwen3.8 Max (2.4T-A95B), Claude Opus 4.8, GPT-5.4 here, though they rank higher overall. Trails GPT-5.6 Luna here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
FrontierMath tier 4 v2	56.1% ✎	▁▂▁▂█▅▃▂▁▁▁▁ beats 48/63	very hard · avg 0.1%	▇▇▇▇▇▇▇▇ 20%	sharp · at its level
FrontierMath tiers 1–3 v2	78.9% ✎	▂··▁▁▃▄▄█▅▃▃ beats 67/82	very hard · avg 1.2%	▇▇▇▇▇▇▇ 19%	sharp · at its level
OTIS mock AIME 2024–2025	98.9% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 178/193	very hard · avg 8.5%	▇ 0.6%	sharp · maxed out · noisy

other tests9 not in a task · 28% of its score

test	result	vs the others in its run	difficulty	counts	why
Furniture Assembly (Epoch)	44.2% ✎	█▆▆▆▆▂▅▂▂▂·▅ beats 19/30	very hard · avg 1.2%	▇▇▇ 7.9%	sharp · at its level
Mystery Game Puzzles (Epoch)	7.0% ✎	█··▂▃▇▄█▂▃▁▁ beats 0/74	very hard · avg 9.4%	▇▇▇ 6.9%	sharp · at its level · noisy
Chess Puzzles (Epoch)	31.0% ✎	█▁▁▁▁▂▁▄▃▂▁▁ beats 105/135	very hard · avg 5.9%	▇▇ 4.1%	sharp · at its level · noisy
GPQA Diamond	90.5% ✎	▁▁▁▁▂▄▆▄▅▇█▅ beats 175/208	hard · avg 40%	▇ 2.9%	sharp
SimpleQA Verified	41.4% ✎	▁▁▃▂▃▃▆▆█▃▂▅ beats 38/78	very hard · avg 8.0%	▇ 1.8%	at its level
Icelandic LLM Leaderboard average (Miðeind, Sept 2026 suite)	57.1% ✎	▁▁▃▂▂▄▁▃▄▄▄█ beats 25/38	very hard · avg 14%	▇ 1.6%	at its level
HealthBench Hard lab	25.4% ✎	▂··▃····▂▄·█ beats 1/8	very hard · avg 9.7%	▇ 1.0%	blunt · at its level
HealthBench lab	50.0% ✎	▃·▃·····█▃▃▅ beats 1/8	hard · avg 24%	▇ 1.0%	blunt · at its level

1 more test

test	result	vs the others in its run	difficulty	counts	why
RobloxQA v2.0 multiple choice (Torpedo Software)	95.6% ✎	▁▁▃▁█▄▃▃▆▂▂▂ beats 30/49	easy · avg 78%	▇ 1.0%	maxed out

$ where to run it

Through an API

api · $0.20/M · $0.1 in / $0.5 out per 1M tokens · 1.1M context · platform.openai.com ↗ · price sourceStandard base token rate from models.dev. Provider-specific tiers, long-context surcharges, caching and promotions may differ.

$ how the score was worked out · Epoch AI — internal evaluatio… 63%, LMArena leaderboard dataset 21%, Terminal-Bench 4.0 official l… 5% and 3 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
Epoch AI — internal evaluations	8	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 63%	CC BY 4.0
LMArena leaderboard dataset	3	▇▇▇▇▇▇ 21%	CC BY 4.0
Terminal-Bench 4.0 official leaderboard submissions	1	▇▇ 5%	Apache-2.0
OpenAI vendor-reported benchmark scores for its own models (openai.com announcement posts and Deployment Safety Hub system cards) · lab-reported	3	▇ 5%	Factual vendor-reported score (not licensed content)
Roblox LLM Leaderboard (Torpedo Software): RobloxQA v2.0 and Luau Leetcode	2	▇ 4%	MIT
Miðeind Icelandic LLM Leaderboard	1	▇ 2%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 76.8