LLM Leaderboard 2026 — Compare 188 AI Models Across 150 Benchmarks

Unified Model Leaderboard

Benchmarks, pricing, runtime signals, and context window in one table. Filter state syncs to the URL so every view is shareable. Provisional-ranked mode includes source-unverified non-generated benchmark evidence.

Methodology Providers AI Race

CSV JSON

188 models

Provisional-ranked mode includes source-unverified non-generated benchmark evidence.Score confidence:Full sourced coverageGood sourced coverageLimited sourced coverageEstimated


1 Claude Mythos Preview Anthropic	Anthropic	Closed	Current	Reasoning	1M	$25.00 / $125.00	N/A	N/A	99	100	100	—	98	99	100	90	—	—
2 GPT-5.4 OpenAI	OpenAI	Closed	Current	Reasoning	1.05M	$2.50 / $15.00	74	151.79s	94	94	91	93	88	98	100	94	95	1465.79
3 Gemini 3.1 Pro Google	Google	Closed	Current	Standard	1M	$1.25 / $5.00	109	29.71s	94	88	94	97	90	96	100	93	71	1492.63
4 Claude Opus 4.6 Anthropic	Anthropic	Closed	Current	Standard	1M	$15.00 / $75.00	40	1.78s	92	93	91	90	84	92	100	95	89	1496.61
5 GPT-5.4 Pro OpenAI	OpenAI	Closed	Current	Reasoning	1.05M	$30.00 / $180.00	74	151.79s	92	92	93	99	100	60	—	94	100	1483.56
6 GPT-5.3 Codex OpenAI	OpenAI	Closed	Current	Reasoning	400K	$2.50 / $10.00	79	88.26s	~89	86	88	95	95	94	100	91	100	1416
7 Gemini 3 Pro Deep Think Google	Google	Closed	Current	Reasoning	2M	N/A	N/A	N/A	~87	88	77	89	100	89	85	83	96	1486.39
8 Claude Sonnet 4.6 Anthropic	Anthropic	Closed	Current	Standard	200K	$3.00 / $15.00	44	1.48s	86	85	83	83	95	85	91	82	78	1462.21
9 GLM-5 (Reasoning) Z.AI	Z.AI	Open	Current	Reasoning	200K	$0.00 / $0.00	N/A	N/A	~85	86	76	88	73	84	82	81	93	1455.62
10 GLM-5.1 Z.AI	Z.AI	Open	Current	Reasoning	203K	$1.40 / $4.40	N/A	N/A	84	83	83	65	—	85	—	93	89	1467.44
11 GPT-5.2 OpenAI	OpenAI	Closed	Current	Reasoning	400K	$2.00 / $8.00	73	130.34s	~84	66	84	86	86	93	99	86	84	1439.54
12 Gemini 3 Pro Google	Google	Closed	Current	Standard	2M	N/A	109	32.65s	~83	76	75	82	86	84	82	79	84	1486.16
13 Grok 4.1 xAI	xAI	Closed	Superseded	Standard	1M	$3.00 / $15.00	N/A	N/A	~81	73	69	92	98	95	100	86	92	1460.98
14 Qwen3.5 397B (Reasoning) Alibaba	Alibaba	Open	Current	Reasoning	128K	$0.00 / $0.00	N/A	N/A	~81	77	85	82	59	80	86	82	92	1450
15 GPT-5.1 OpenAI	OpenAI	Closed	Current	Reasoning	200K	$1.50 / $6.00	111	57.47s	~81	81	81	68	96	84	86	78	70	1438.53
16 Claude Opus 4.5 Anthropic	Anthropic	Closed	Current	Standard	200K	N/A	46	1.01s	80	81	79	70	72	84	84	58	95	1468
17 GPT-5 (high) OpenAI	OpenAI	Closed	Established	Reasoning	128K	N/A	83	36.28s	~80	82	73	78	92	81	82	83	72	1433.37
18 GPT-5.2-Codex OpenAI	OpenAI	Closed	Current	Reasoning	400K	$2.00 / $8.00	123	87.34s	~80	84	81	89	89	80	88	93	98	1331
19 Kimi K2.5 (Reasoning) Moonshot AI	Moonshot AI	Closed	Current	Reasoning	128K	N/A	N/A	N/A	~79	69	87	70	71	75	90	100	68	1447
20 GPT-5.1-Codex-Max OpenAI	OpenAI	Closed	Current	Reasoning	400K	$2.00 / $8.00	N/A	N/A	~79	81	79	90	90	81	86	89	97	1349
21 Grok 4.20 xAI	xAI	Closed	Current	Reasoning	2M	$2.00 / $6.00	233	10.33s	78	64	80	69	68	—	—	98	—	1490.38
22 GLM-5 Z.AI	Z.AI	Open	Superseded	Standard	200K	$0.00 / $0.00	74	1.64s	77	73	77	63	56	85	73	81	89	1455.57
23 Qwen3.6 Plus Alibaba	Alibaba	Closed	Current	Reasoning	1M	$0.00 / $0.00	N/A	N/A	77	72	80	44	74	77	82	90	—	—
24 Gemma 4 31B Google	Google	Open	Current	Reasoning	256K	$0.00 / $0.00	N/A	N/A	~74	—	87	55	71	75	—	—	—	1451.16
25 GPT-5 (medium) OpenAI	OpenAI	Closed	Established	Reasoning	128K	N/A	83	36.28s	~74	75	81	75	89	76	87	78	92	1328

Showing 25 of 188

Scoring Methodology

Each model's overall score is a normalized weighted average of category averages. Within each category, benchmark scores are normalized to a common scale and combined using per-benchmark weights that favor harder, less-saturated evaluations.

Each score includes a confidence indicator (1-4 dots) showing how much sourced benchmark data supports it — models with no non-generated benchmark coverage are marked as estimated.

Display-only benchmarks like MMLU, HumanEval, BBH, LisanBench, FLTEval, and the AIME/HMMT exams remain visible but are excluded from scoring.

Data sourced from OpenBench, official model papers, and public leaderboards. External consensus signals are used as bounded calibration inputs but are not exposed in exported data.

Compare frontier AI models by quality, cost, and context

Decision-ready picks

Unified Model Leaderboard

The AI models change fast. We track them for you.

Rankings

Popular Comparisons

Tools & Resources

Scoring Methodology

Stay ahead of the LLM curve