Where do these scores come from?

Every score comes from a public benchmark or official machine-readable file, with a source link and date. This site does not invent scores, and missing values are not treated as zero.

How is the overall score calculated?

Each source is min-max normalized within itself, then weighted: Artificial Analysis 28%, LMArena Text Arena 28%, LiveBench 22%, coding composite 15%, reasoning composite 7%. Missing sources are dropped and the rest are renormalized. A model needs at least two headline sources (or three partials) to enter the overall board. When several variants of one family exist, only the best overall variant is shown.

Are site prices part of the benchmarks?

No. Benchmark scores come from third-party public boards. CNY prices are this site’s catalogue reference rates (CNY 1 = 1,000 points) and are settled on actual usage. They are shown separately so billing is not confused with quality.

How often is this updated?

A collector runs automatically every day. The upstream boards may not publish a new snapshot daily, so each page shows both this site’s collection time and the source publication date.

88Z.ai / RANKINGS

Overall AI model leaderboard

A weighted composite of public benchmarks. Missing scores are omitted and weights are renormalized; they are never filled with zeros.

Last collected by this site: 2026-11-15 06:40 (UTC+8, Asia/Shanghai). Source publication dates are in the table or the citations below.

A weighted composite of public benchmarks. Missing scores are omitted and weights are renormalized; they are never filled with zeros.

138 model families (best published variant per family)

#	Model	Overall	AA	LMArena	LiveBench	Coding	Reasoning	Our price / 1M tokens
1	Claude Fable 5.1Anthropic	96.5	—	—	84.0	98.2	87.3	in ¥86 / out ¥430
2	GPT 6 AstraOpenAI	96.5	—	—	84.5	95.0	91.0	in ¥86 / out ¥430
3	Gemini 4 CarbonGoogle	96.0	—	—	83.5	100	83.2	in ¥17.2 / out ¥86
4	Claude Opus 5.5Anthropic	95.0	—	—	83.4	94.5	89.2	in ¥34.4 / out ¥172
5	GPT 6.1 SolOpenAI	91.6	—	—	81.4	87.4	95.3	in ¥17.2 / out ¥86
6	Claude Sonnet 5.5 (xHigh)Anthropic	91.1	—	1,473	—	91.5	92.2	in ¥17.2 / out ¥86
7	Claude Fable 5.2Anthropic	90.6	—	—	84.9	83.3	76.5	in ¥68.8 / out ¥344
8	Gemini 4 Hydrogen (High)Google	90.4	—	1,490	—	82.6	93.6	in ¥12.9 / out ¥64.5
9	Claude Opus 5 (Max)Anthropic	90.2	—	1,489	—	80.9	96.6	in ¥43 / out ¥215
10	GPT 6.1 TerraOpenAI	89.5	—	—	78.8	97.5	76.1	in ¥8.6 / out ¥43
11	Muse Spark 1.3 (Max)Meta	89.3	—	1,497	—	76.2	95.3	in ¥10.75 / out ¥36.55
12	Claude Fable 5 (High)Anthropic	89.2	—	1,504	—	72.3	97.5	in ¥86 / out ¥430
13	Gemini 4 ArgonGoogle	89.0	—	—	77.7	92.5	90.0	in ¥17.2 / out ¥86
14	GPT 5 (High)OpenAI	88.9	—	1,435	—	100	84.3	in ¥10.75 / out ¥86
15	Kimi K3 (Max)Moonshot	88.4	—	1,487	—	76.5	94.7	in ¥1.65 / out ¥137.6
16	Claude Opus 4.6 (High)Anthropic	85.9	—	1,503	—	62.1	96.8	in ¥43 / out ¥215
17	Gemini 3.7 Flash (High)Google	85.9	—	1,489	—	67.5	94.6	in ¥6.45 / out ¥32.25
18	Claude Opus 4.7 (High)Anthropic	85.9	—	1,501	—	63.4	94.9	in ¥43 / out ¥215
19	DeepSeek v4.1 Flash 20260910 (Max)DeepSeek	85.6	—	1,475	—	71.7	93.6	—
20	Gemini 3 FlashGoogle	85.5	—	1,473	—	73.3	90.4	—
21	Gemini 3 ProGoogle	85.3	—	1,486	—	68.8	90.4	—
22	GLM 5.2 (Max)zai	84.9	—	1,475	—	70.0	92.3	in ¥0.06 / out ¥58.48
23	GPT 6 Sol (Max)OpenAI	84.7	—	1,457	—	78.5	85.1	in ¥17.2 / out ¥86
24	Muse Spark 1.1Meta	84.0	—	1,491	—	61.4	93.3	in ¥10.75 / out ¥36.55
25	Muse Spark 1.2 (xHigh)Meta	83.8	—	1,493	—	60.4	92.2	in ¥10.75 / out ¥36.55
26	Claude Opus 4.8 (High)Anthropic	83.7	—	1,482	—	63.2	93.2	in ¥43 / out ¥215
27	Gemini 3.6 Flash (High)Google	83.2	—	1,483	—	60.9	94.5	in ¥6.45 / out ¥32.25
28	Grok 4.6 (High)xAI	82.6	—	1,457	—	71.4	85.8	in ¥17.2 / out ¥51.6
29	GLM 5Z-AI	82.5	—	1,457	—	70.8	85.5	in ¥8.6 / out ¥27.52
30	DeepSeek V4 Pro High 20260813DeepSeek	82.0	—	1,464	—	66.5	86.5	—
31	Grok 4.5xAI	81.5	—	1,466	—	62.8	89.8	in ¥17.2 / out ¥51.6
32	Gemini 3.5 Flash (High)Google	81.3	—	1,478	—	55.9	94.0	in ¥12.9 / out ¥77.4
33	Claude Sonnet 4.6Anthropic	80.8	—	1,472	—	58.9	88.4	in ¥25.8 / out ¥129
34	DeepSeek R1 0528DeepSeek	80.7	—	1,422	—	80.3	77.7	in ¥4.91 / out ¥19.66
35	Claude Sonnet 5 (High)Anthropic	80.5	—	1,461	—	61.1	90.1	in ¥17.2 / out ¥86
36	Claude Opus 4.5 High 32k 20251101Anthropic	80.0	—	1,474	—	55.1	90.1	—
37	GPT 5.2OpenAI	79.8	—	1,435	—	69.9	83.8	in ¥15.05 / out ¥120.4
38	Gemini 3.1 Pro Preview--	79.8	—	1,487	—	49.0	92.5	in ¥17.2 / out ¥103.2
39	GLM 5.1zai	79.8	—	1,465	—	57.0	90.4	in ¥12.04 / out ¥37.84
40	Kimi k2.6Moonshot	79.5	—	1,461	—	57.2	91.1	in ¥5.59 / out ¥29.33
41	Mimo v2.5 Proxiaomi	78.9	—	1,466	—	53.3	91.1	in ¥3.74 / out ¥7.48
42	Mimo v2.6 Flash小米集团	78.9	50.3	1,448	—	74.0	59.2	in ¥1.2 / out ¥2.41
43	Hy3--	78.8	—	1,454	—	56.7	91.9	in ¥1.14 / out ¥4.54
44	Claude Opus 4.5 20251101Anthropic	78.4	—	1,470	—	51.9	88.5	—
45	Claude Haiku 5.5 (High)Anthropic	78.3	—	1,421	—	68.2	86.7	in ¥0.86 / out ¥4.3
46	qwen3.6 Max PreviewAlibaba	78.2	—	1,460	—	53.7	89.9	—
47	Solar Pro 4$0.96	77.9	63.3	1,387	—	39.0	81.1	in ¥1.29 / out ¥5.16
48	DeepSeek v4.1 Flash 1017DeepSeek	77.6	—	—	70.3	85.0	78.8	in ¥0.33 / out ¥4.3
49	GPT 5.6 TerraOpenAI	77.1	—	—	67.2	90.4	83.2	in ¥17.2 / out ¥103.2
50	GPT 5.1OpenAI	76.9	—	1,438	—	59.8	82.5	in ¥10.75 / out ¥86
51	GPT 5.4OpenAI	76.9	—	1,465	—	48.5	88.0	in ¥6.45 / out ¥38.7
52	DeepSeek V4 Pro Preview (High)DeepSeek	76.8	—	1,455	—	51.2	88.9	—
53	DeepSeek v3.2DeepSeek	76.2	—	1,425	—	61.6	82.8	in ¥2.24 / out ¥3.27
54	qwen3.8 Max 0902阿里巴巴	76.1	48.6	—	—	79.0	70.0	in ¥17.2 / out ¥51.6
55	DeepSeek V4 ProDeepSeek	76.1	—	1,458	—	49.3	85.5	in ¥7.74 / out ¥12.9
56	Muse Wonder 1.0Meta	76.0	—	—	70.7	77.2	83.2	—
57	GLM 4.5Z.ai	76.0	—	1,411	—	66.7	80.6	in ¥5.16 / out ¥18.92
58	Kimi k2.5 (Thinking)Moonshot	75.4	—	1,450	—	47.7	89.6	in ¥4.3 / out ¥21.5
59	qwen3.6 PlusAlibaba	75.3	—	1,443	—	51.0	86.8	in ¥2.8 / out ¥16.77
60	Grok 4.20 Beta Reasoning 0309xAI	75.1	—	1,472	—	39.8	88.8	—
61	qwen3.8 (Max)Alibaba	75.0	—	1,484	60.9	67.8	84.7	—
62	Mimo V2 Proxiaomi	74.7	—	1,448	—	47.4	86.6	—
63	MiniMax m2.5MiniMax	74.6	—	1,391	—	69.8	77.7	in ¥2.32 / out ¥8.17
64	Claude Sonnet 4.5 High 32k 20250929Anthropic	74.0	—	1,456	—	42.1	87.4	—
65	Beam 1.0 501b A23breflection	73.9	—	1,434	—	47.7	89.9	in ¥2.58 / out ¥8.6
66	Gemini 3 Flash MinimalGoogle	73.8	—	1,458	—	40.9	87.1	—
67	DeepSeek R1DeepSeek	73.7	—	1,398	—	63.2	80.7	—
68	GLM 4.7zai	73.6	—	1,441	—	47.5	83.0	in ¥3.44 / out ¥15.05
69	DeepSeek V4 Flash Preview (High)DeepSeek	73.5	—	1,439	—	47.0	85.2	—
70	Mimo v2.5xiaomi	73.2	—	1,434	—	47.9	84.6	in ¥3.44 / out ¥17.2
71	GLM 4.6Z.ai	73.1	—	1,424	—	52.0	81.6	in ¥4.3 / out ¥17.2
72	GPT 5.4 Mini (High)OpenAI	73.0	—	1,447	—	42.7	84.9	in ¥6.45 / out ¥38.7
73	Claude Sonnet 4.5 20250929Anthropic	73.0	—	1,455	—	41.0	83.0	—
74	Gemini 2.5 ProGoogle	72.9	—	1,445	—	43.2	84.9	in ¥10.75 / out ¥86
75	qwen3.5 397b A17bAlibaba	72.7	—	1,441	—	43.0	86.6	in ¥5.16 / out ¥30.96
76	Claude Opus 4.1 20250805Anthropic	72.6	—	1,448	—	41.6	84.0	—
77	DeepSeek V3 0324DeepSeek	71.8	—	1,395	—	61.0	73.7	in ¥2.49 / out ¥9.8
78	Kimi k2.5 InstantMoonshot	71.7	—	1,431	—	43.7	85.2	—
79	GLM 5.3 FlashZhipu	71.6	45.5	1,475	60.7	70.9	53.3	in ¥0.34 / out ¥0.67
80	O3 Mini (High)OpenAI	71.4	—	1,364	—	67.3	79.6	—
81	Gemma 4 26b A4bGoogle	71.1	—	1,437	—	37.6	89.6	in ¥0.58 / out ¥1.94
82	Grok 4.3xAI	70.4	—	1,442	—	37.4	81.8	in ¥10.75 / out ¥21.5
83	MiniMax m2.7MiniMax	69.4	—	1,415	—	42.7	82.4	in ¥1.81 / out ¥7.22
84	Nvidia Nemotron 3.5 Super Vl 120b A12b Nvfp4NVIDIA	69.4	—	1,414	—	43.1	81.4	—
85	Muse GlimmerMeta	69.1	—	1,425	—	37.0	85.4	—
86	Grok 4.7xAI	69.1	—	—	67.9	74.7	61.9	in ¥17.2 / out ¥51.6
87	GLM 5.3 (Max)$5.01	69.0	25	1,482	—	71.9	94.6	in ¥0.72 / out ¥30.96
88	Mimo v2.6 Pro小米集团	68.6	35.7	1,480	66.1	67.4	44.7	in ¥3.74 / out ¥7.48
89	Kimi K2 Thinking TurboMoonshot	68.2	—	1,430	—	32.7	84.4	—
90	qwen3.5 122b A10bAlibaba	68.2	—	1,416	—	37.8	82.7	in ¥2.24 / out ¥17.89
91	Grok 4.1 (Thinking)xAI	67.7	—	1,465	—	19.0	85.5	—
92	Hy3 PreviewTencent	67.5	—	1,409	—	37.5	83.4	in ¥1.55 / out ¥5.16
93	qwen3.5 27bAlibaba	67.3	—	1,408	—	37.4	82.4	in ¥2.24 / out ¥22.36
94	Qwen3 Coder 480b A35b InstructAlibaba	67.2	—	1,387	—	47.5	75.1	—
95	Claude Haiku 4.5 20251001Anthropic	66.3	—	1,414	—	33.9	79.1	—
96	Inkling 2 Small$1.18	66.1	30.9	1,412	—	75.2	—	in ¥2.58 / out ¥8.6
97	Gemini 3.1 Flash Lite PreviewGoogle	66.0	—	1,433	—	24.2	84.4	in ¥2.15 / out ¥12.9
98	qwen3.8 27bAlibaba	65.8	—	1,434	54.1	68.5	65.2	in ¥1.29 / out ¥16.13
99	Kimi k3.1 ProMoonshot	65.7	39.7	—	68.7	67.2	67.5	in ¥17.2 / out ¥86
100	MiniMax m2.1 PreviewMiniMax	65.2	—	1,384	—	40.9	77.2	—
101	Mistral Large 4 0Mistral	65.1	—	—	60.3	83.6	63.1	in ¥6.02 / out ¥18.06
102	Muse Wonder (Max)$6.95	65.1	14.3	1,511	—	75.2	92.8	in ¥34.4 / out ¥172
103	DeepSeek v3.2 ExpDeepSeek	64.9	—	1,422	—	26.1	80.6	in ¥2.32 / out ¥3.53
104	Grok 4.1 Fast ReasoningxAI	64.8	—	1,431	—	22.3	82.0	—
105	Gemini 2.5 FlashGoogle	64.5	—	1,409	—	29.1	79.6	in ¥2.58 / out ¥21.5
106	MiniMax M2MiniMax	64.2	—	1,344	—	53.3	71.8	—
107	Mimo V2 Flash Non Thinking (Thinking)xiaomi	63.6	—	1,392	—	34.0	75.0	—
108	Gemini 3.8 Flash (High)$7.77	63.2	14.6	1,497	—	66.9	97.3	in ¥7.1 / out ¥35.48
109	Mistral Large 3Mistral	62.5	—	1,415	—	21.1	79.1	—
110	Qwen3 32bAlibaba	62.4	—	1,347	—	43.1	78.6	in ¥0.69 / out ¥2.41
111	Mimo V2 Flash (Thinking)xiaomi	61.4	—	1,386	—	28.9	74.6	—
112	qwen3.5 35b A3bAlibaba	61.2	—	1,394	—	23.8	78.3	in ¥1.2 / out ¥8.6
113	qwen3.5 FlashAlibaba	61.1	—	1,396	—	22.5	79.0	in ¥0.56 / out ¥2.24
114	Grok 4 Fast ReasoningxAI	59.1	—	1,405	—	12.0	80.5	—
115	GPT 6 LunaOpenAI	59.0	—	—	60.4	71.4	50.4	in ¥0.86 / out ¥4.3
116	Trinity Large Thinking--	57.6	—	1,367	—	21.6	76.2	—
117	GPT Oss 120bOpenAI	57.1	—	1,352	—	25.1	75.9	in ¥1.29 / out ¥5.16
118	Inklingthinky	55.2	—	1,442	44.3	42.5	64.2	in ¥8.6 / out ¥34.83
119	Mistral Large 4Mistral	54.0	—	1,420	40.9	49.2	73.1	in ¥6.02 / out ¥18.06
120	Qwq 32bAlibaba	53.7	—	1,336	—	20.5	73.1	—
121	Step 5 Preview阶跃星辰	53.5	14.3	1,449	54.9	66.9	59.2	in ¥8.6 / out ¥23.22
122	MiniMax M3稀宇科技	53.4	13.5	1,436	—	53.9	66.3	in ¥2.58 / out ¥10.32
123	GPT 4o 2024 08 06OpenAI	53.4	—	1,336	—	23.1	64.7	in ¥21.5 / out ¥86
124	Gemma 3 27b ItGoogle	50.1	—	1,365	—	1.5	66.9	in ¥0.69 / out ¥3.01
125	Yi LightningOther	48.9	—	1,328	—	11.0	64.0	—
126	Granite 4.1 8bIBM	48.3	—	1,304	—	16.0	66.1	—
127	Gemini 3.5 Flash Lite$0.69	48.2	1.7	1,455	—	47.1	85.4	in ¥2.58 / out ¥21.5
128	MiniMax m3.1 Flash稀宇科技	47.3	16.6	—	63.7	67.8	57.2	in ¥1.29 / out ¥5.16
129	Inkling Small$0.20	46.8	5.5	1,405	—	44.0	86.9	in ¥3.87 / out ¥10.32
130	GPT 4o Mini 2024 07 18OpenAI	43.9	—	1,318	—	0	59.5	in ¥1.29 / out ¥5.16
131	Mistral Medium 3.5$1.04	43.5	4.5	1,426	—	25.1	82.9	in ¥12.9 / out ¥64.5
132	Gemma 4 31bGoogle	42.2	—	1,452	32.4	19.5	44.6	in ¥1.72 / out ¥3.44
133	qwen2.5 Coder 32b InstructAlibaba	41.5	—	1,271	—	7.6	58.6	in ¥5.68 / out ¥8.6
134	Granite 4.2 8b$0.0405	37.7	12	1,289	—	—	—	in ¥0.52 / out ¥2.15
135	Muse Spark OpenMeta	35.3	—	—	48.5	49.2	20.1	in ¥10.75 / out ¥36.55
136	Nemotron Super Vl 3.5NVIDIA	33.9	—	—	47.0	47.8	23.5	—
137	Smaug Nemotron 3.5 Nvfp4NVIDIA	33.0	—	—	47.0	44.1	25.4	—
138	Nemotron Super Vl 3.5 Nvfp4NVIDIA	18.4	—	—	42.3	20.5	12.7	—

Sources

LMArena Text Arena (style-controlled) — Official Arena leaderboard dataset, text_style_control / latest. Scores are Arena ratings (Bradley-Terry).
LMArena WebDev Arena — Official webdev / latest snapshot, measuring front-end and web-app battles.
LMArena Vision Arena (style-controlled) — Official vision_style_control / latest snapshot.
LMArena Text-to-Image Arena — Official text_to_image / latest snapshot.
LMArena Text-to-Video Arena — Official text_to_video / latest snapshot.
LMArena Agent Arena — Official agent / latest aggregate scores.
Artificial Analysis Intelligence Index — Intelligence Index, cost per task, time per task and first-token latency from the official public leaderboard page.
LiveBench 2.0 — Official CSV results. Overall uses the published formula: Long Agentic ×2, Design ×1.5, Coding ×1.5, other suites ×1, then a weighted mean over suites that have a score.
Aider Polyglot — Official polyglot_leaderboard.yml. The metric is pass_rate_2 on 225 cross-language exercises. Repeated submissions keep the best score.
SWE-bench Verified (mini-SWE-agent) — Official leaderboards.json, Verified board, resolved % under the mini-SWE-agent protocol so models are compared in a similar scaffold. Scores from other agents are not mixed into the composite coding score.
SuperCLUE Chinese general board — Total and section scores from the official monthly general-board spreadsheet.

FAQ

Where do these scores come from?: Every score comes from a public benchmark or official machine-readable file, with a source link and date. This site does not invent scores, and missing values are not treated as zero.
How is the overall score calculated?: Each source is min-max normalized within itself, then weighted: Artificial Analysis 28%, LMArena Text Arena 28%, LiveBench 22%, coding composite 15%, reasoning composite 7%. Missing sources are dropped and the rest are renormalized. A model needs at least two headline sources (or three partials) to enter the overall board. When several variants of one family exist, only the best overall variant is shown.
Are site prices part of the benchmarks?: No. Benchmark scores come from third-party public boards. CNY prices are this site’s catalogue reference rates (CNY 1 = 1,000 points) and are settled on actual usage. They are shown separately so billing is not confused with quality.
How often is this updated?: A collector runs automatically every day. The upstream boards may not publish a new snapshot daily, so each page shows both this site’s collection time and the source publication date.