Research · Benchmark profile

AA Long Context Reasoning

Artificial Analysis long-context reasoning benchmark measuring extraction and synthesis over documents from roughly 10k to 100k tokens.

Ranking weightedCurrent / rollingActive95 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on AA Long Context Reasoning

Leaderboard · 95 models

GPT-5.2 Codex leads at 75.7%, followed by GPT-5 (75.6%) and GPT-5.1 (75%).

Ranking · 1.5%ResearchCurrent

#1Closed

GPT-5.2 Codex

OpenAI

gpt-5.2-codex

75.7%

Overall #77 · Context 400K tokens

#2Closed

GPT-5

OpenAI

gpt-5

75.6%

Overall #63 · Context 400K tokens

#3Closed

GPT-5.1

OpenAI

gpt.5.1

75%

Overall #32 · Context Unavailable

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

OpenAIMoonshot AIAnthropicMiniMaxXiaomiGoogleZ.AI

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
GPT-5.2 Codex	#1	OpenAI	75.7%
GPT-5	#2	OpenAI	75.6%
GPT-5.1	#3	OpenAI	75%
Kimi K3	#4	Moonshot AI	74.7%
GPT-5.5	#5	OpenAI	74.3%
Claude Opus 4.5	#6	Anthropic	74%
GPT-5.3-Codex	#7	OpenAI	74%
GPT-5.4	#8	OpenAI	74%
GPT-5.6 Luna	#9	OpenAI	74%
GPT-5.6 Terra	#10	OpenAI	74%
MiniMax M3	#11	MiniMax	74%
GPT-5.6 Sol	#12	OpenAI	73.7%
MiMo-V2.5-Pro	#13	Xiaomi	73.3%
Gemini 3.1 Pro Preview	#14	Google	72.7%
GPT-5.2	#15	OpenAI	72.7%
GLM-5.2	#16	Z.AI	71.3%
Claude Opus 4.6	#17	Anthropic	70.7%
Claude Sonnet 4.6	#18	Anthropic	70.7%
Claude Sonnet 5	#19	Anthropic	70.7%
Gemini 3 Pro	#20	Google	70.7%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
GPT-5.2 Codex	#1	OpenAI	75.7%
GPT-5	#2	OpenAI	75.6%
GPT-5.1	#3	OpenAI	75%
Kimi K3	#4	Moonshot AI	74.7%
GPT-5.5	#5	OpenAI	74.3%
Claude Opus 4.5	#6	Anthropic	74%
GPT-5.3-Codex	#7	OpenAI	74%
GPT-5.4	#8	OpenAI	74%
GPT-5.6 Luna	#9	OpenAI	74%
GPT-5.6 Terra	#10	OpenAI	74%
MiniMax M3	#11	MiniMax	74%
GPT-5.6 Sol	#12	OpenAI	73.7%
MiMo-V2.5-Pro	#13	Xiaomi	73.3%
Gemini 3.1 Pro Preview	#14	Google	72.7%
GPT-5.2	#15	OpenAI	72.7%
GLM-5.2	#16	Z.AI	71.3%
Claude Opus 4.6	#17	Anthropic	70.7%
Claude Sonnet 4.6	#18	Anthropic	70.7%
Claude Sonnet 5	#19	Anthropic	70.7%
Gemini 3 Pro	#20	Google	70.7%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	1	74.7%
OpenAI	10	74.3%
MiniMax	1	74%
Xiaomi	1	73.3%
Google	2	71.7%
Anthropic	4	71.5%
Z.AI	1	71.3%

Benchmark score table (95 models)

One best score per model · higher is better

AA Long Context Reasoning leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	GPT-5.2 Codex gpt-5.2-codex	OpenAI	closed	75.7%
#2	GPT-5 gpt-5	OpenAI	closed	75.6%
#3	GPT-5.1 gpt.5.1	OpenAI	closed	75%
#4	Kimi K3 kimi-k3	Moonshot AI	closed	74.7%
#5	GPT-5.5 gpt-5.5	OpenAI	closed	74.3%
#6	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	74%
#7	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	74%
#8	GPT-5.4 gpt-5.4	OpenAI	closed	74%
#9	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	74%
#10	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	74%
#11	MiniMax M3 MiniMax-M3	MiniMax	—	74%
#12	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	73.7%
#13	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	73.3%
#14	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	72.7%
#15	GPT-5.2 gpt-5.2	OpenAI	closed	72.7%
#16	GLM-5.2 glm-5-2	Z.AI	open	71.3%
#17	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	70.7%
#18	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	70.7%
#19	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	70.7%
#20	Gemini 3 Pro gemini-3-pro	Google	closed	70.7%
#21	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	70.3%
#22	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	70.3%
#23	Claude Fable 5 claude-fable-5	Anthropic	closed	70%
#24	Kimi K2.6 kimi-k2.6	Moonshot AI	open	69.7%
#25	Muse Spark muse-spark	Meta	closed	69.7%
#26	Qwen3.6 Max qwen3.6-max	Alibaba Cloud	closed	69.7%
#27	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	69.7%
#28	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	69.3%
#29	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	69.3%
#30	o3 o3	OpenAI	closed	69.3%
#31	GPT-5 Codex gpt-5-codex	OpenAI	closed	69%
#32	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	69%
#33	MiniMax M2.7 minimax-m2-7	MiniMax	open	68.7%
#34	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	68.7%
#35	Grok 4 grok-4	xAI	closed	68%
#36	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	67.7%
#37	Grok 4.5 grok-4.5	xAI	closed	67.7%
#38	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	67.3%
#39	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	67%
#40	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	66.7%
#41	Qwen3.5 122B qwen3.5-122b-a10b	Alibaba Cloud	open	66.7%
#42	Claude Opus 4.1 claude-opus-4-1	Anthropic	closed	66.3%
#43	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	66.3%
#44	Gemini 3 Flash gemini-3-flash	Google	closed	66.3%
#45	Kimi K2 Thinking kimi-k2-thinking	Moonshot AI	open	66.3%
#46	Kimi K2.7 Code kimi-k2.7-code	Moonshot AI	open	66.3%
#47	Gemini 2.5 Pro gemini-2.5-pro	Google	closed	66%
#48	GPT-5 mini gpt-5-mini	OpenAI	closed	66%
#49	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	66%
#50	MiniMax M2.5 minimax-m2.5	MiniMax	open	66%
#51	Qwen3 Max qwen3-max	Alibaba Cloud	closed	66%
#52	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	65.7%
#53	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	65.7%
#54	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	65.3%
#55	Kimi K2.5 kimi-k2.5	Moonshot AI	—	65.3%
#56	DeepSeek V3.1 Terminus deepseek-v3.1-terminus	DeepSeek	open	65%
#57	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	65%
#58	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	65%
#59	Claude Sonnet 4 claude-sonnet-4	Anthropic	closed	64.7%
#60	Grok 4 Fast grok-4-fast	xAI	closed	64.7%
#61	Gemini 2.5 Flash gemini-2.5-flash	Google	closed	64.3%
#62	Grok 4.3 grok-4.3	xAI	closed	64.3%
#63	GLM-4.7 glm-4.7	Z.AI	open	64%
#64	GLM-5 glm-5	Z.AI	open	63.3%
#65	Muse Spark 1.1 muse-spark-1-1	Meta	closed	63.3%
#66	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	63%
#67	MiMo-V2-Flash mimo-v2-flash	Xiaomi	open	63%
#68	MiMo-V2.5 mimo-v2.5	Xiaomi	open	62.7%
#69	Qwen3.5 35B qwen3.5-35b-a3b	Alibaba Cloud	open	62.7%
#70	GLM-5.1 glm-5-1	Z.AI	open	62.3%
#71	Gemma 4 31B gemma-4-31b	Google	open	62%
#72	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	61%
#73	MiniMax M2 minimax-m2	MiniMax	open	61%
#74	Mistral Medium 3.5 mistral-medium-3-5	Mistral AI	open	61%
#75	Claude Sonnet 3.7 claude-3-7-sonnet	Anthropic	closed	60.7%
#76	GLM-5-Turbo glm-5-turbo	Z.AI	closed	60.7%
#77	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	60.7%
#78	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	60%
#79	o1 o1	OpenAI	closed	59.3%
#80	MiniMax M2.1 minimax-m2.1	MiniMax	open	59%
#81	Grok 4.20 grok-4.20	xAI	closed	58%
#82	Gemma 4 26B gemma-4-26b-a4b	Google	open	55.7%
#83	Gemma 4 12B gemma-4-12b	Google	open	55.3%
#84	o4-mini o4-mini	OpenAI	closed	55%
#85	GLM-4.6 glm-4.6	Z.AI	open	54.3%
#86	Nova 2 Pro nova-2-pro-preview	Amazon	closed	54.3%
#87	Qwen3.5 Omni Plus qwen3.5-omni-plus	Alibaba Cloud	closed	52.7%
#88	Kimi K2 0905 kimi-k2-0905	Moonshot AI	open	52.3%
#89	Grok 3 mini grok-3-mini	xAI	closed	50.3%
#90	Grok Code Fast 1 grok-code-fast-1	xAI	closed	48.3%
#91	Command A+ command-a-plus	Cohere	open	46%
#92	Mistral Small 4 mistral-small-2603	Mistral AI	open	44.7%
#93	Ling 2.6 1T ling-2.6-1t	InclusionAI	open	34.7%
#94	Mistral Large 3 mistral-large-2512	Mistral AI	open	34.7%
#95	Claude Opus 4 claude-opus-4	Anthropic	closed	33.7%

The top of this snapshot is led by GPT-5.2 Codex at 75.7%; third place is 0.67 points behind. The top-10 spread is 1.7 points.

Compare top models

GPT-5.2 Codex vs GPT-5 GPT-5 vs GPT-5.1 GPT-5.1 vs Kimi K3 Kimi K3 vs GPT-5.5

About AA Long Context Reasoning

Definition and scoring

Organisation: Artificial Analysis
Category: Research
Version: Not evaluated
Direction: higher is better
Ranking use: Weighted 1.5%
Contamination risk: Low

Artificial Analysis long-context reasoning benchmark measuring extraction and synthesis over documents from roughly 10k to 100k tokens. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does AA Long Context Reasoning measure?

Artificial Analysis long-context reasoning benchmark measuring extraction and synthesis over documents from roughly 10k to 100k tokens.

Which model scores highest on AA Long Context Reasoning?

GPT-5.2 Codex by OpenAI currently leads with 75.7%.

How many models are evaluated on AA Long Context Reasoning?

95 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.5% of its capability category. GPT-5.2 Codex is overall #77.

MMLU-Pro

Rolling · 22 results · Weighted 3.5%

SimpleQA

Rolling · 2 results · Weighted 3.5%

PaperBench

Rolling · 2 results · Weighted 3.5%

Finance Agent v2

v2 · 4 results · Not Weighted

BenchLM Knowledge prior

bench-align-v5.1 · 33 results · Weighted 1.6%

AA-Omniscience

Rolling · 94 results · Weighted 1.5%

Research · Benchmark profile

AA Long Context Reasoning

Artificial Analysis long-context reasoning benchmark measuring extraction and synthesis over documents from roughly 10k to 100k tokens.

Ranking weightedCurrent / rollingActive95 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on AA Long Context Reasoning

Leaderboard · 95 models

GPT-5.2 Codex leads at 75.7%, followed by GPT-5 (75.6%) and GPT-5.1 (75%).

Ranking · 1.5%ResearchCurrent

#1Closed

GPT-5.2 Codex

OpenAI

gpt-5.2-codex

75.7%

Overall #77 · Context 400K tokens

#2Closed

GPT-5

OpenAI

gpt-5

75.6%

Overall #63 · Context 400K tokens

#3Closed

GPT-5.1

OpenAI

gpt.5.1

75%

Overall #32 · Context Unavailable

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

OpenAIMoonshot AIAnthropicMiniMaxXiaomiGoogleZ.AI

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
GPT-5.2 Codex	#1	OpenAI	75.7%
GPT-5	#2	OpenAI	75.6%
GPT-5.1	#3	OpenAI	75%
Kimi K3	#4	Moonshot AI	74.7%
GPT-5.5	#5	OpenAI	74.3%
Claude Opus 4.5	#6	Anthropic	74%
GPT-5.3-Codex	#7	OpenAI	74%
GPT-5.4	#8	OpenAI	74%
GPT-5.6 Luna	#9	OpenAI	74%
GPT-5.6 Terra	#10	OpenAI	74%
MiniMax M3	#11	MiniMax	74%
GPT-5.6 Sol	#12	OpenAI	73.7%
MiMo-V2.5-Pro	#13	Xiaomi	73.3%
Gemini 3.1 Pro Preview	#14	Google	72.7%
GPT-5.2	#15	OpenAI	72.7%
GLM-5.2	#16	Z.AI	71.3%
Claude Opus 4.6	#17	Anthropic	70.7%
Claude Sonnet 4.6	#18	Anthropic	70.7%
Claude Sonnet 5	#19	Anthropic	70.7%
Gemini 3 Pro	#20	Google	70.7%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
GPT-5.2 Codex	#1	OpenAI	75.7%
GPT-5	#2	OpenAI	75.6%
GPT-5.1	#3	OpenAI	75%
Kimi K3	#4	Moonshot AI	74.7%
GPT-5.5	#5	OpenAI	74.3%
Claude Opus 4.5	#6	Anthropic	74%
GPT-5.3-Codex	#7	OpenAI	74%
GPT-5.4	#8	OpenAI	74%
GPT-5.6 Luna	#9	OpenAI	74%
GPT-5.6 Terra	#10	OpenAI	74%
MiniMax M3	#11	MiniMax	74%
GPT-5.6 Sol	#12	OpenAI	73.7%
MiMo-V2.5-Pro	#13	Xiaomi	73.3%
Gemini 3.1 Pro Preview	#14	Google	72.7%
GPT-5.2	#15	OpenAI	72.7%
GLM-5.2	#16	Z.AI	71.3%
Claude Opus 4.6	#17	Anthropic	70.7%
Claude Sonnet 4.6	#18	Anthropic	70.7%
Claude Sonnet 5	#19	Anthropic	70.7%
Gemini 3 Pro	#20	Google	70.7%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	1	74.7%
OpenAI	10	74.3%
MiniMax	1	74%
Xiaomi	1	73.3%
Google	2	71.7%
Anthropic	4	71.5%
Z.AI	1	71.3%

Benchmark score table (95 models)

One best score per model · higher is better

AA Long Context Reasoning leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	GPT-5.2 Codex gpt-5.2-codex	OpenAI	closed	75.7%
#2	GPT-5 gpt-5	OpenAI	closed	75.6%
#3	GPT-5.1 gpt.5.1	OpenAI	closed	75%
#4	Kimi K3 kimi-k3	Moonshot AI	closed	74.7%
#5	GPT-5.5 gpt-5.5	OpenAI	closed	74.3%
#6	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	74%
#7	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	74%
#8	GPT-5.4 gpt-5.4	OpenAI	closed	74%
#9	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	74%
#10	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	74%
#11	MiniMax M3 MiniMax-M3	MiniMax	—	74%
#12	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	73.7%
#13	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	73.3%
#14	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	72.7%
#15	GPT-5.2 gpt-5.2	OpenAI	closed	72.7%
#16	GLM-5.2 glm-5-2	Z.AI	open	71.3%
#17	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	70.7%
#18	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	70.7%
#19	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	70.7%
#20	Gemini 3 Pro gemini-3-pro	Google	closed	70.7%
#21	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	70.3%
#22	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	70.3%
#23	Claude Fable 5 claude-fable-5	Anthropic	closed	70%
#24	Kimi K2.6 kimi-k2.6	Moonshot AI	open	69.7%
#25	Muse Spark muse-spark	Meta	closed	69.7%
#26	Qwen3.6 Max qwen3.6-max	Alibaba Cloud	closed	69.7%
#27	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	69.7%
#28	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	69.3%
#29	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	69.3%
#30	o3 o3	OpenAI	closed	69.3%
#31	GPT-5 Codex gpt-5-codex	OpenAI	closed	69%
#32	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	69%
#33	MiniMax M2.7 minimax-m2-7	MiniMax	open	68.7%
#34	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	68.7%
#35	Grok 4 grok-4	xAI	closed	68%
#36	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	67.7%
#37	Grok 4.5 grok-4.5	xAI	closed	67.7%
#38	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	67.3%
#39	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	67%
#40	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	66.7%
#41	Qwen3.5 122B qwen3.5-122b-a10b	Alibaba Cloud	open	66.7%
#42	Claude Opus 4.1 claude-opus-4-1	Anthropic	closed	66.3%
#43	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	66.3%
#44	Gemini 3 Flash gemini-3-flash	Google	closed	66.3%
#45	Kimi K2 Thinking kimi-k2-thinking	Moonshot AI	open	66.3%
#46	Kimi K2.7 Code kimi-k2.7-code	Moonshot AI	open	66.3%
#47	Gemini 2.5 Pro gemini-2.5-pro	Google	closed	66%
#48	GPT-5 mini gpt-5-mini	OpenAI	closed	66%
#49	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	66%
#50	MiniMax M2.5 minimax-m2.5	MiniMax	open	66%
#51	Qwen3 Max qwen3-max	Alibaba Cloud	closed	66%
#52	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	65.7%
#53	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	65.7%
#54	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	65.3%
#55	Kimi K2.5 kimi-k2.5	Moonshot AI	—	65.3%
#56	DeepSeek V3.1 Terminus deepseek-v3.1-terminus	DeepSeek	open	65%
#57	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	65%
#58	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	65%
#59	Claude Sonnet 4 claude-sonnet-4	Anthropic	closed	64.7%
#60	Grok 4 Fast grok-4-fast	xAI	closed	64.7%
#61	Gemini 2.5 Flash gemini-2.5-flash	Google	closed	64.3%
#62	Grok 4.3 grok-4.3	xAI	closed	64.3%
#63	GLM-4.7 glm-4.7	Z.AI	open	64%
#64	GLM-5 glm-5	Z.AI	open	63.3%
#65	Muse Spark 1.1 muse-spark-1-1	Meta	closed	63.3%
#66	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	63%
#67	MiMo-V2-Flash mimo-v2-flash	Xiaomi	open	63%
#68	MiMo-V2.5 mimo-v2.5	Xiaomi	open	62.7%
#69	Qwen3.5 35B qwen3.5-35b-a3b	Alibaba Cloud	open	62.7%
#70	GLM-5.1 glm-5-1	Z.AI	open	62.3%
#71	Gemma 4 31B gemma-4-31b	Google	open	62%
#72	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	61%
#73	MiniMax M2 minimax-m2	MiniMax	open	61%
#74	Mistral Medium 3.5 mistral-medium-3-5	Mistral AI	open	61%
#75	Claude Sonnet 3.7 claude-3-7-sonnet	Anthropic	closed	60.7%
#76	GLM-5-Turbo glm-5-turbo	Z.AI	closed	60.7%
#77	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	60.7%
#78	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	60%
#79	o1 o1	OpenAI	closed	59.3%
#80	MiniMax M2.1 minimax-m2.1	MiniMax	open	59%
#81	Grok 4.20 grok-4.20	xAI	closed	58%
#82	Gemma 4 26B gemma-4-26b-a4b	Google	open	55.7%
#83	Gemma 4 12B gemma-4-12b	Google	open	55.3%
#84	o4-mini o4-mini	OpenAI	closed	55%
#85	GLM-4.6 glm-4.6	Z.AI	open	54.3%
#86	Nova 2 Pro nova-2-pro-preview	Amazon	closed	54.3%
#87	Qwen3.5 Omni Plus qwen3.5-omni-plus	Alibaba Cloud	closed	52.7%
#88	Kimi K2 0905 kimi-k2-0905	Moonshot AI	open	52.3%
#89	Grok 3 mini grok-3-mini	xAI	closed	50.3%
#90	Grok Code Fast 1 grok-code-fast-1	xAI	closed	48.3%
#91	Command A+ command-a-plus	Cohere	open	46%
#92	Mistral Small 4 mistral-small-2603	Mistral AI	open	44.7%
#93	Ling 2.6 1T ling-2.6-1t	InclusionAI	open	34.7%
#94	Mistral Large 3 mistral-large-2512	Mistral AI	open	34.7%
#95	Claude Opus 4 claude-opus-4	Anthropic	closed	33.7%

The top of this snapshot is led by GPT-5.2 Codex at 75.7%; third place is 0.67 points behind. The top-10 spread is 1.7 points.

Compare top models

GPT-5.2 Codex vs GPT-5 GPT-5 vs GPT-5.1 GPT-5.1 vs Kimi K3 Kimi K3 vs GPT-5.5

About AA Long Context Reasoning

Definition and scoring

Organisation: Artificial Analysis
Category: Research
Version: Not evaluated
Direction: higher is better
Ranking use: Weighted 1.5%
Contamination risk: Low

Open benchmark source ↗

FAQ

Quick answers

What does AA Long Context Reasoning measure?

Artificial Analysis long-context reasoning benchmark measuring extraction and synthesis over documents from roughly 10k to 100k tokens.

Which model scores highest on AA Long Context Reasoning?

GPT-5.2 Codex by OpenAI currently leads with 75.7%.

How many models are evaluated on AA Long Context Reasoning?

95 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.5% of its capability category. GPT-5.2 Codex is overall #77.

MMLU-Pro

Rolling · 22 results · Weighted 3.5%

SimpleQA

Rolling · 2 results · Weighted 3.5%

PaperBench

Rolling · 2 results · Weighted 3.5%

Finance Agent v2

v2 · 4 results · Not Weighted

BenchLM Knowledge prior

bench-align-v5.1 · 33 results · Weighted 1.6%

AA-Omniscience

Rolling · 94 results · Weighted 1.5%

Leaderboard · 95 models

GPT-5.2 Codex

GPT-5

GPT-5.1

See the field, not just the table.

Benchmark score table (95 models)

Compare top models

Definition and scoring

Quick answers

What does AA Long Context Reasoning measure?

Which model scores highest on AA Long Context Reasoning?

How many models are evaluated on AA Long Context Reasoning?

Does this affect overall Lumina rank?

More Research benchmarks

Leaderboard · 95 models

GPT-5.2 Codex

GPT-5

GPT-5.1

See the field, not just the table.

Benchmark score table (95 models)

Compare top models

Definition and scoring

Quick answers

What does AA Long Context Reasoning measure?

Which model scores highest on AA Long Context Reasoning?

How many models are evaluated on AA Long Context Reasoning?

Does this affect overall Lumina rank?

More Research benchmarks