LMArena TextBenchmark scores and sources

Published result; benchmark version and evaluation conditions remain in the id and result note.

pillar: preference · weight 1 within pillar · unit: elo (higher is better) · official board ↗

Row	Model	Result	Normalized (0–100)	Confidence
1	Claude Opus 5.5 Anthropic	1514.8 elo	82.8	100% confidence 100 percent, Full
2	Claude Fable 5.1 Anthropic	1510.0 elo	82.5	100% confidence 100 percent, Full
3	Claude Opus 5 Anthropic	1502.6 elo	82.0	100% confidence 100 percent, Full
4	Gemini 3.8 Flash Google	1499.0 elo	81.7	100% confidence 100 percent, Full
5	Claude Opus 4.6 Anthropic	1498.1 elo	81.6	100% confidence 100 percent, Full
6	MiMo-V2.6-Pro Xiaomi	1491.7 elo	81.1	88% confidence 88 percent, High
7	Claude Fable 5 Anthropic	1491.2 elo	81.1	100% confidence 100 percent, Full
8	Muse Spark 1.3 Meta	1489.6 elo	81.0	93% confidence 93 percent, High
9	Gemini 3.7 Flash Google	1486.0 elo	80.7	100% confidence 100 percent, Full
10	Qwen3.8 Max Alibaba / Qwen	1483.3 elo	80.5	85% confidence 85 percent, High
11	Claude Opus 4.7 Anthropic	1483.2 elo	80.5	100% confidence 100 percent, Full
12	Muse Spark 1.2 Meta	1482.3 elo	80.4	53% confidence 53 percent, Medium
13	Gemini 3.1 Pro Preview Google	1480.5 elo	80.3	100% confidence 100 percent, Full
14	Gemini 3.6 Flash Google	1478.8 elo	80.1	100% confidence 100 percent, Full
15	Muse Spark 1.1 Meta	1478.6 elo	80.1	61% confidence 61 percent, Medium
16	Gemini 3.5 Flash Google	1477.5 elo	80.0	100% confidence 100 percent, Full
17	Kimi K3 Moonshot AI	1475.5 elo	79.9	100% confidence 100 percent, Full
18	GLM-5.3 Z.ai	1471.2 elo	79.5	93% confidence 93 percent, High
19	Claude Sonnet 5.5 Anthropic	1471.2 elo	79.5	93% confidence 93 percent, High
20	GLM-5.3-Flash Z.ai	1471.1 elo	79.5	100% confidence 100 percent, Full
21	GLM-5.2 Z.ai	1469.8 elo	79.4	100% confidence 100 percent, Full
22	GPT-5.5 OpenAI	1466.8 elo	79.2	100% confidence 100 percent, Full
23	MiMo-V2.5-Pro Xiaomi	1465.3 elo	79.0	88% confidence 88 percent, High
24	DeepSeek V4.1 Flash DeepSeek	1462.4 elo	78.8	69% confidence 69 percent, Medium
25	GLM-5.1 Z.ai	1461.4 elo	78.7	64% confidence 64 percent, Medium
26	Claude Sonnet 4.6 Anthropic	1457.7 elo	78.4	100% confidence 100 percent, Full
27	Gemini 2.5 Pro Google	1457.7 elo	78.4	90% confidence 90 percent, High
28	GPT-5.6 Sol OpenAI	1457.3 elo	78.4	100% confidence 100 percent, Full
29	Kimi K2.6 Moonshot AI	1455.3 elo	78.2	85% confidence 85 percent, High
30	MiMo-V2.6-Flash Xiaomi	1455.2 elo	78.2	88% confidence 88 percent, High
31	Qwen3.7 Plus Alibaba / Qwen	1454.8 elo	78.1	64% confidence 64 percent, Medium
32	Claude Opus 4.8 Anthropic	1453.8 elo	78.1	100% confidence 100 percent, Full
33	GPT-5.4 OpenAI	1452.1 elo	77.9	100% confidence 100 percent, Full
34	DeepSeek V4 Pro DeepSeek	1451.1 elo	77.8	85% confidence 85 percent, High
35	Grok 4.20 (Reasoning) xAI	1450.8 elo	77.8	80% confidence 80 percent, High
36	Claude Opus 4.5 Anthropic	1448.3 elo	77.6	100% confidence 100 percent, Full
37	Grok 4.5 xAI	1448.1 elo	77.6	100% confidence 100 percent, Full
38	Qwen3.6 Max Preview Alibaba / Qwen	1446.9 elo	77.5	64% confidence 64 percent, Medium
39	GPT-5.6 Terra OpenAI	1446.5 elo	77.4	100% confidence 100 percent, Full
40	GPT-6.1 Sol OpenAI	1446.5 elo	77.4	100% confidence 100 percent, Full
41	GLM-5 Z.ai	1446.2 elo	77.4	90% confidence 90 percent, High
42	Kimi K2.5 Moonshot AI	1445.2 elo	77.3	100% confidence 100 percent, Full
43	Gemma 4 31B IT Google	1443.0 elo	77.1	64% confidence 64 percent, Medium
44	Gemini 3 Flash Preview Google	1442.4 elo	77.1	93% confidence 93 percent, High
45	Claude Sonnet 5 Anthropic	1442.1 elo	77.0	93% confidence 93 percent, High
46	Step 5 Preview StepFun	1441.7 elo	77.0	75% confidence 75 percent, Medium
47	Inkling Thinking Machines Lab	1441.3 elo	77.0	100% confidence 100 percent, Full
48	Qwen3.8 27B Alibaba / Qwen	1441.1 elo	76.9	69% confidence 69 percent, Medium
49	GPT-6 Astra OpenAI	1440.5 elo	76.9	100% confidence 100 percent, Full
50	GLM-4.6 Z.ai	1439.8 elo	76.8	55% confidence 55 percent, Medium
51	Hy3 Tencent	1438.8 elo	76.7	88% confidence 88 percent, High
52	Qwen3.5 397B-A17B Alibaba / Qwen	1437.9 elo	76.7	69% confidence 69 percent, Medium
53	GLM-5V-Turbo Z.ai	1437.2 elo	76.6	32% confidence 32 percent, Low
54	Qwen3.6 Plus Alibaba / Qwen	1436.6 elo	76.5	80% confidence 80 percent, High
55	MiMo-V2-Pro Xiaomi	1436.0 elo	76.5	75% confidence 75 percent, Medium
56	GLM-4.7 Z.ai	1435.2 elo	76.4	69% confidence 69 percent, Medium
57	Gemini 3.5 Flash Lite Google	1434.9 elo	76.4	100% confidence 100 percent, Full
58	Claude Sonnet 4.5 Anthropic	1434.5 elo	76.4	80% confidence 80 percent, High
59	Gemma 4 26B A4B IT Google	1434.2 elo	76.3	64% confidence 64 percent, Medium
60	MiniMax-M3 MiniMax	1432.6 elo	76.2	100% confidence 100 percent, Full
61	DeepSeek V4 Flash DeepSeek	1432.1 elo	76.1	53% confidence 53 percent, Medium
62	GPT-5.6 Luna OpenAI	1430.5 elo	76.0	100% confidence 100 percent, Full
63	GLM-4.5 Z.ai	1430.0 elo	76.0	47% confidence 47 percent, Low
64	Grok 4.6 xAI	1427.8 elo	75.8	100% confidence 100 percent, Full
65	Mistral Large 3 Mistral AI	1427.6 elo	75.7	37% confidence 37 percent, Low
66	MiMo-V2.5 Xiaomi	1427.6 elo	75.7	88% confidence 88 percent, High
67	Mistral Large 4 Mistral AI	1427.4 elo	75.7	64% confidence 64 percent, Medium
68	MiMo-V2-Omni Xiaomi	1423.1 elo	75.3	75% confidence 75 percent, Medium
69	GPT-5.1 OpenAI	1422.2 elo	75.2	85% confidence 85 percent, High
70	Mistral Medium 3.5 Mistral AI	1420.7 elo	75.1	37% confidence 37 percent, Low
71	Qwen3 VL 235B A22B Instruct Alibaba / Qwen	1420.1 elo	75.0	32% confidence 32 percent, Low
72	DeepSeek V3.2 DeepSeek	1420.0 elo	75.0	56% confidence 56 percent, Medium
73	DeepSeek-V3.1 DeepSeek	1419.7 elo	75.0	32% confidence 32 percent, Low
74	Qwen3 235B-A22B Instruct 2507 Alibaba / Qwen	1419.2 elo	74.9	48% confidence 48 percent, Low
75	GPT-5.5 Instant OpenAI	1419.0 elo	74.9	64% confidence 64 percent, Medium
76	Claude Opus 4.1 Anthropic	1418.6 elo	74.9	80% confidence 80 percent, High
77	Qwen3.5 122B-A10B Alibaba / Qwen	1417.3 elo	74.8	37% confidence 37 percent, Low
78	Qwen3-Next 80B-A3B Instruct Alibaba / Qwen	1417.0 elo	74.7	32% confidence 32 percent, Low
79	Gemini 2.5 Flash Google	1416.9 elo	74.7	47% confidence 47 percent, Low
80	Gemini 3.1 Flash Lite Preview Google	1415.8 elo	74.6	48% confidence 48 percent, Low
81	Kimi K2 Thinking Turbo Moonshot AI	1414.5 elo	74.5	64% confidence 64 percent, Medium
82	Inkling Small Thinking Machines Lab	1413.7 elo	74.4	100% confidence 100 percent, Full
83	Qwen3 Max Alibaba / Qwen	1412.8 elo	74.3	69% confidence 69 percent, Medium
84	GPT-5.2 OpenAI	1412.2 elo	74.3	100% confidence 100 percent, Full
85	GPT-5.4 mini OpenAI	1411.5 elo	74.2	100% confidence 100 percent, Full
86	o3 OpenAI	1409.9 elo	74.1	87% confidence 87 percent, High
87	Qwen3.5 27B Alibaba / Qwen	1408.9 elo	74.0	37% confidence 37 percent, Low
88	Grok 4.1 Fast (Reasoning) xAI	1408.0 elo	73.9	32% confidence 32 percent, Low
89	GPT-5 OpenAI	1406.2 elo	73.7	100% confidence 100 percent, Full
90	MiniMax-M2.7 MiniMax	1404.6 elo	73.6	88% confidence 88 percent, High
91	Step 3.5 Flash StepFun	1402.9 elo	73.4	88% confidence 88 percent, High
92	Qwen3 VL 235B A22B Thinking Alibaba / Qwen	1400.4 elo	73.1	32% confidence 32 percent, Low
93	Grok 4.7 xAI	1399.3 elo	73.0	100% confidence 100 percent, Full
94	Qwen3.5 Flash Alibaba / Qwen	1397.0 elo	72.8	64% confidence 64 percent, Medium
95	Grok 4.3 xAI	1396.9 elo	72.8	80% confidence 80 percent, High
96	Claude Haiku 4.5 Anthropic	1396.4 elo	72.7	64% confidence 64 percent, Medium
97	GPT-6 Sol OpenAI	1395.0 elo	72.6	100% confidence 100 percent, Full
98	MiMo-V2-Flash Xiaomi	1394.7 elo	72.6	75% confidence 75 percent, Medium
99	Qwen3.5 35B-A3B Alibaba / Qwen	1394.5 elo	72.6	64% confidence 64 percent, Medium
100	GPT-6 Luna OpenAI	1391.1 elo	72.2	100% confidence 100 percent, Full
101	GPT-5.3 Chat (latest) OpenAI	1389.1 elo	72.0	32% confidence 32 percent, Low
102	Solar Pro 4 Upstage	1386.2 elo	71.7	100% confidence 100 percent, Full
103	GLM-4.5-Air Z.ai	1383.7 elo	71.5	37% confidence 37 percent, Low
104	GPT-4.1 OpenAI	1382.9 elo	71.4	100% confidence 100 percent, Full
105	Nemotron 3 Super 120B A12B NVIDIA	1377.7 elo	70.9	100% confidence 100 percent, Full
106	GLM-4.6V Z.ai	1376.7 elo	70.8	32% confidence 32 percent, Low
107	DeepSeek V3 0324 DeepSeek	1374.6 elo	70.5	74% confidence 74 percent, Medium
108	GPT-5 Mini OpenAI	1373.3 elo	70.4	99% confidence 99 percent, High
109	DeepSeek-R1 DeepSeek	1372.7 elo	70.3	88% confidence 88 percent, High
110	GPT-5.4 nano OpenAI	1372.1 elo	70.3	100% confidence 100 percent, Full
111	Qwen3-Next 80B-A3B (Thinking) Alibaba / Qwen	1368.1 elo	69.9	32% confidence 32 percent, Low
112	Qwen Max Alibaba / Qwen	1366.7 elo	69.7	75% confidence 75 percent, Medium
113	Qwen3 235B-A22B Alibaba / Qwen	1366.3 elo	69.7	76% confidence 76 percent, Medium
114	Claude Opus 4 Anthropic	1365.8 elo	69.6	100% confidence 100 percent, Full
115	GPT OSS 120B OpenAI	1365.4 elo	69.6	79% confidence 79 percent, Medium
116	Nova 2 Lite Amazon	1361.6 elo	69.2	100% confidence 100 percent, Full
117	MiniMax-M2.5 MiniMax	1359.4 elo	68.9	100% confidence 100 percent, Full
118	Gemma 3 27B IT Google	1357.7 elo	68.8	74% confidence 74 percent, Medium
119	Qwen3-Coder 480B-A35B Instruct Alibaba / Qwen	1356.5 elo	68.6	47% confidence 47 percent, Low
120	o4-mini OpenAI	1353.2 elo	68.3	87% confidence 87 percent, High
121	GLM-4.7-Flash Z.ai	1350.7 elo	68.0	69% confidence 69 percent, Medium
122	GPT-4.1 mini OpenAI	1340.4 elo	66.9	87% confidence 87 percent, High
123	Qwen3 32B Alibaba / Qwen	1340.1 elo	66.8	67% confidence 67 percent, Medium
124	Trinity Large Thinking Arcee AI	1340.0 elo	66.8	100% confidence 100 percent, Full
125	MiniMax-M2 MiniMax	1339.9 elo	66.8	96% confidence 96 percent, High
126	Claude Sonnet 4 Anthropic	1339.3 elo	66.7	100% confidence 100 percent, Full
127	Trinity Large Preview Arcee AI	1338.8 elo	66.7	100% confidence 100 percent, Full
128	Llama 3.3 Nemotron Super 49B v1.5 NVIDIA	1338.1 elo	66.6	100% confidence 100 percent, Full
129	Mistral Small 3.2 Mistral AI	1338.1 elo	66.6	32% confidence 32 percent, Low
130	Gemma 3 12B IT Google	1334.2 elo	66.2	64% confidence 64 percent, Medium
131	DeepSeek-V3 DeepSeek	1332.5 elo	66.0	93% confidence 93 percent, High
132	GLM-4.5V Z.ai	1332.5 elo	66.0	37% confidence 37 percent, Low
133	Command A Cohere	1330.9 elo	65.8	75% confidence 75 percent, Medium
134	QwQ 32B Alibaba / Qwen	1329.1 elo	65.6	67% confidence 67 percent, Medium
135	GPT-5 Nano OpenAI	1319.7 elo	64.5	85% confidence 85 percent, High
136	o3-mini OpenAI	1319.4 elo	64.5	96% confidence 96 percent, High
137	Qwen3 30B A3B Alibaba / Qwen	1316.8 elo	64.2	64% confidence 64 percent, Medium
138	GPT-4o (2024-05-13) OpenAI	1300.4 elo	62.3	93% confidence 93 percent, High
139	Claude Sonnet 3.7 Anthropic	1299.4 elo	62.2	100% confidence 100 percent, Full
140	Claude Sonnet 3.5 v2 Anthropic	1297.8 elo	62.0	100% confidence 100 percent, Full
141	Gemma 3 4B IT Google	1290.8 elo	61.2	64% confidence 64 percent, Medium
142	GPT OSS 20B OpenAI	1287.4 elo	60.7	64% confidence 64 percent, Medium
143	GPT-4o mini OpenAI	1286.4 elo	60.6	100% confidence 100 percent, Full
144	GPT-4.1 nano OpenAI	1284.8 elo	60.4	82% confidence 82 percent, High
145	Llama 3.1 Nemotron 70B Instruct NVIDIA	1282.6 elo	60.2	100% confidence 100 percent, Full
146	GPT-4o (2024-08-06) OpenAI	1282.5 elo	60.2	100% confidence 100 percent, Full
147	Llama 4 Scout 17B Instruct Meta	1279.2 elo	59.8	64% confidence 64 percent, Medium
148	Mistral Small 3.1 24B Mistral AI	1277.3 elo	59.5	64% confidence 64 percent, Medium
149	Llama-3.3-70B-Instruct Meta	1274.2 elo	59.2	100% confidence 100 percent, Full
150	Mistral Large 2.1 Mistral AI	1265.3 elo	58.1	100% confidence 100 percent, Full
151	Llama-3.1-70B-Instruct Meta	1260.9 elo	57.6	93% confidence 93 percent, High
152	Nova Pro Amazon	1258.7 elo	57.3	100% confidence 100 percent, Full
153	Claude Haiku 3.5 Anthropic	1255.4 elo	56.9	100% confidence 100 percent, Full
154	Qwen2.5-Coder-32B-Instruct Alibaba / Qwen	1230.0 elo	53.7	90% confidence 90 percent, High
155	Command R+ Cohere	1228.9 elo	53.6	75% confidence 75 percent, Medium
156	Nova Lite Amazon	1228.5 elo	53.6	100% confidence 100 percent, Full
157	Aya Expanse 32B Cohere	1224.0 elo	53.0	75% confidence 75 percent, Medium
158	Nova Micro Amazon	1208.4 elo	51.1	100% confidence 100 percent, Full
159	Claude Haiku 3 Anthropic	1194.7 elo	49.3	93% confidence 93 percent, High
160	Command R Cohere	1187.3 elo	48.4	75% confidence 75 percent, Medium
161	Llama-3.1-8B-Instruct Meta	1186.6 elo	48.3	93% confidence 93 percent, High
162	Aya Expanse 8B Cohere	1185.1 elo	48.1	75% confidence 75 percent, Medium
163	Mistral Medium 3 Mistral AI	1165.1 elo	45.7	85% confidence 85 percent, High
164	Llama-3.2-3B Meta	1109.5 elo	38.9	75% confidence 75 percent, Medium
165	Llama-3.2-1B Meta	1054.6 elo	32.6	93% confidence 93 percent, High

Results are as published by the source behind each value (hover or tap the number). Benchmark scores are shown individually for every source/variant (a model may have multiple rows), and vary by version, harness and date; the normalized column uses the method’s fixed 0–100 scales and feeds the preference pillar of the SI Score.

LMArena TextBenchmark scores and sources

What changed

Alerts on this device