LMArena Hard prompts leaderboard

LMArena’s text votes on the hardest prompts: specific, complex and needing real problem solving. Gemini 4 Argon leads with a rating of 1538.

Models: 351
Leader: Gemini 4 Argon
Source: LMArena
As of: Oct 22, 2026

LMArena Hard prompts, rating against blended price per million tokens

Open weightsProprietaryBest for the price

Full leaderboard

Rank	Model	Lab	Rating	95% interval	Votes	Price in / out
1	Gemini 4 Argon	Google	1538	1530–1546	6.3K	$2 / $10
2	Claude Opus 5.5	Anthropic	1536	1528–1545	6.4K	$4 / $20
3	Claude Fable 5.1	Anthropic	1528	1521–1535	9.5K	$10 / $50
4	Claude Opus 5	Anthropic	1527	1521–1532	22.2K	$5 / $25
5	Claude Opus 4.6	Anthropic	1524	1520–1528	53.1K	$5 / $25
8	MiMo-V2.6-Pro	Xiaomi	1511	1499–1522	2.6K	$0.43 / $0.87
9	Gemini 3.8 Flash	Google	1509	1503–1514	28.3K	$0.82 / $4.13
10	Claude Fable 5	Anthropic	1507	1502–1512	28.7K	$10 / $50
11	Claude Opus 4.7	Anthropic	1506	1501–1510	43.7K	$5 / $25
12	Muse Wonder	Meta	1505	1495–1515	3.7K	$4 / $20
13	Muse Spark 1.3	Meta	1504	1498–1510	13.3K	$1.25 / $4.25
15	Gemini 3.7 Flash	Google	1497	1492–1503	19.8K	$0.75 / $3.75
16	Qwen3.8-Max	Alibaba	1497	1491–1502	20.1K	$2 / $6
17	Kimi K3	Moonshot AI	1495	1490–1500	22.6K	$0.24 / $11
18	Claude Sonnet 5.5	Anthropic	1493	1485–1502	5.5K	$2 / $10
19	GLM-5.3-Flash	Z.ai	1491	1486–1497	24.4K	$0.15 / $0.50
20	Muse Spark 1.2	Meta	1490	1482–1498	6.7K	$1.25 / $4.25
21	GLM-5.3	Z.ai	1490	1484–1496	15.8K	$0.079 / $1.40
22	GPT-5.5	OpenAI	1489	1485–1493	45.5K	$5 / $30
23	Gemini 3.5 Flash	Google	1488	1483–1492	32.4K	$1.50 / $9
25	Gemini 3.6 Flash	Google	1486	1480–1491	24.1K	$0.75 / $3.75
26	MiMo-V2.5-Pro	Xiaomi	1486	1481–1490	50.7K	$0.30 / $0.61
27	Muse Spark 1.1	Meta	1485	1481–1490	30.4K	$1.25 / $4.25
28	GPT-5.4	OpenAI	1485	1481–1490	42.1K	$2.50 / $15
30	Gemini 3.1 Pro Preview	Google	1485	1481–1488	84.9K	$2 / $12
31	Claude Sonnet 4.6	Anthropic	1485	1480–1489	46.2K	$3 / $15
32	GPT-5.6 Sol	OpenAI	1484	1479–1489	27.7K	$4 / $20
33	DeepSeek-V4.1-Flash-20260910	DeepSeek	1483	1476–1490	8.8K	$0.15 / $0.60
34	Qwen3.5-Max-Preview	Alibaba	1483	1477–1488	14.5K	—
35	Qwen3.7-Max-Preview	Alibaba	1482	1471–1494	2.7K	$1.48 / $4.42
36	Claude Opus 4.8	Anthropic	1482	1478–1487	42.5K	$5 / $25
37	ERNIE 5.1	Baidu	1481	1475–1486	25.7K	—
38	GLM-5.2	Z.ai	1481	1476–1485	33.9K	$0.034 / $7
39	Gemini 3 Pro	Google	1480	1476–1485	23.2K	—
40	Claude Opus 4.5	Anthropic	1476	1472–1480	43.2K	$5 / $25
41	Muse Spark	Meta	1474	1468–1481	9.2K	—
43	MiMo-V2.6-Flash	Xiaomi	1474	1466–1481	8.4K	$0.14 / $0.28
45	GLM-5.1	Z.ai	1473	1469–1477	42.7K	$1.40 / $4.40
46	Kimi K2.6	Moonshot AI	1470	1465–1475	26.1K	$0.65 / $3.41
47	GPT-5.6 Terra	OpenAI	1469	1465–1474	30.2K	$2 / $12
49	GPT-6.1 Sol	OpenAI	1465	1456–1474	4.8K	$2 / $10
50	Gemini 3 Flash	Google	1465	1459–1470	17.2K	$0.50 / $3
51	GPT-6 Astra	OpenAI	1463	1456–1471	8.5K	$10 / $50
52	Step 5 Preview	StepFun	1463	1455–1471	6.1K	$1 / $2.70
53	Grok 4.5	xAI	1462	1457–1467	26.6K	$2 / $6
54	Claude Sonnet 4.5	Anthropic	1462	1458–1465	48.3K	$3 / $15
56	DeepSeek-V4-Pro	DeepSeek	1461	1457–1466	38.5K	$0.66 / $1.98
57	Claude Sonnet 5	Anthropic	1461	1456–1466	34K	$2 / $10
58	Qwen3.7-Plus	Alibaba	1460	1455–1465	28.3K	$0.32 / $1.28
59	Qwen3.8-27B	Alibaba	1459	1454–1465	17.1K	$0.20 / $2.50
60	amazon-nova-experimental-chat	Amazon	1459	1447–1471	2K	—
61	MiMo-V2-Pro	Xiaomi	1458	1452–1463	16.5K	—
63	Qwen3.6-Max-Preview	Alibaba	1457	1447–1467	3.6K	—
64	GPT-5.1	OpenAI	1457	1452–1462	22.4K	$1.25 / $10
65	Gemini 2.5 Pro	Google	1455	1452–1458	65.9K	$1.25 / $10
66	Dola Seed 2.0 Pro	ByteDance	1453	1449–1457	51K	—
67	Kimi K2.5	Moonshot AI	1453	1449–1457	47.9K	$0.50 / $2.50
69	NVIDIA-Nemotron-3-Ultra-550B-A55B	NVIDIA	1452	1444–1460	7.6K	—
70	GPT-5.6 Luna	OpenAI	1452	1447–1457	27.2K	$0.20 / $1.20
71	GLM-5	Z.ai	1452	1447–1457	18.2K	$1 / $3.20
72	Grok 4.20 Beta	xAI	1451	1447–1456	43.5K	—
73	Inkling	Thinky	1451	1447–1456	28.6K	$1 / $4.05
74	MiMo-V2.5	Xiaomi	1450	1445–1455	31.9K	$0.12 / $0.24
75	Grok 4.6	xAI	1449	1443–1454	20.8K	$2 / $6
76	Qwen3.6-Plus	Alibaba	1449	1444–1453	32.6K	$0.33 / $1.95
77	Grok 4.20 Multi-Agent Beta	xAI	1448	1444–1453	42.2K	—
78	Qwen3-Max-Preview	Alibaba	1448	1442–1454	13.6K	—
79	Qwen3.5-397B-A17B	Alibaba	1448	1444–1452	63.3K	$0.50 / $3.60
80	Hy3	Tencent	1448	1441–1455	8.7K	$0.083 / $0.33
81	Gemma 4 31B	Google	1448	1438–1458	3.6K	$0.20 / $0.40
82	MiniMax-M3	MiniMax	1447	1442–1451	44.3K	$0.30 / $1.20
83	Mistral Large 4	Mistral AI	1446	1438–1454	6.5K	$1.36 / $4.18
84	ERNIE 5.0	Baidu	1445	1441–1450	21.8K	—
85	GPT-5.2 Chat	OpenAI	1445	1440–1450	22.8K	$1.75 / $14
86	MiMo-V2-Omni	Xiaomi	1445	1438–1451	13.8K	—
87	GLM-5V-Turbo	Z.ai	1444	1436–1451	6.9K	$1.20 / $4
88	DeepSeek-V4-Flash	DeepSeek	1444	1439–1448	34.8K	$0.042 / $0.18
90	GLM-4.7	Z.ai	1443	1435–1451	6.8K	$0.54 / $1.98
91	Claude Opus 4.1	Anthropic	1442	1438–1447	25K	$15 / $75
93	Beam 1.0 501B A23B	Reflection AI	1442	1430–1454	2.7K	$0.25 / $1
94	Grok 4.20 Beta 1	xAI	1441	1435–1446	17.5K	—
95	Gemini 3.5 Flash-Lite	Google	1441	1436–1446	23.8K	$0.30 / $2.50
96	GLM-4.6	Z.ai	1440	1435–1445	19.4K	$0.55 / $2.20
97	LongCat-Flash-Chat Exp	Meituan	1440	1434–1445	18.7K	—
98	Gemma 4 26B A4B	Google	1439	1429–1448	3.5K	$0.068 / $0.23
101	Mistral Medium 3.5	Mistral AI	1436	1429–1444	7.7K	$1.50 / $7.50
102	Grok 4.1	xAI	1435	1432–1439	39.5K	—
104	DeepSeek-V3.2-Exp	DeepSeek	1434	1429–1438	27.1K	$0.28 / $0.42
105	Grok 3 Preview	xAI	1433	1427–1440	10.6K	—
106	Qwen3-235B-A22B-Instruct	Alibaba	1433	1430–1437	53.5K	$0.09 / $0.55
111	Mistral Large 3	Mistral AI	1430	1427–1434	53.1K	$0.50 / $1.50
112	GLM-4.5	Z.ai	1429	1423–1436	11.1K	$0.60 / $2.20
114	GPT-5.2	OpenAI	1428	1424–1433	29.9K	$1.75 / $14
115	Qwen3-VL-235B-A22B-Instruct	Alibaba	1428	1420–1437	5.7K	$0.21 / $1.90
116	Kimi K2 Thinking Turbo	Moonshot AI	1428	1424–1432	37.1K	—
117	Qwen3-Next-80B-A3B-Instruct	Alibaba	1428	1422–1434	12K	$0.09 / $1.10
121	GPT-5.5 Instant	OpenAI	1426	1420–1432	18.4K	—
122	DeepSeek-V3.1-Terminus	DeepSeek	1426	1411–1440	1.7K	$0.27 / $1
123	Mistral Medium 3.1	Mistral AI	1426	1422–1429	52.9K	$0.40 / $2
124	GPT-5.4 mini	OpenAI	1424	1420–1429	41.6K	$0.75 / $4.50
125	ChatGPT-4o Latest	OpenAI	1424	1420–1428	38.4K	—
127	Inkling-Small	Thinky	1423	1418–1428	23K	$0.45 / $1.20
128	MiniMax-M2.7	MiniMax	1423	1419–1426	57K	$0.30 / $1.20
129	Gemini 2.5 Flash	Google	1422	1419–1425	65.1K	$0.30 / $2.50
131	Qwen3.5-122B-A10B	Alibaba	1422	1416–1427	18.8K	$0.40 / $3.20
132	GPT-6 Sol	OpenAI	1421	1414–1428	9.1K	$2 / $10
133	Claude Haiku 5.5	Anthropic	1421	1411–1431	3.7K	$0.10 / $0.50
135	Claude Haiku 4.5	Anthropic	1420	1417–1423	94K	$1 / $5
136	MiMo-V2-Flash	Xiaomi	1420	1416–1424	29.1K	—
137	Hunyuan Vision 1.5 Thinking	Tencent	1417	1400–1434	1.1K	—
138	DeepSeek-V3.1	DeepSeek	1417	1408–1426	5.1K	$0.25 / $0.95
139	Qwen3-235B-A22B-Thinking	Alibaba	1416	1406–1425	3.8K	$0.45 / $3.50
140	GPT-5	OpenAI	1416	1410–1421	15.1K	$1.25 / $10
141	DeepSeek-R1	DeepSeek	1415	1408–1423	6.9K	$0.50 / $2.15
143	Qwen3.5-27B	Alibaba	1414	1409–1420	18.1K	$0.27 / $2.16
145	Grok 4.7	xAI	1414	1407–1421	8.9K	$2 / $6
146	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B	NVIDIA	1414	1402–1425	2.6K	—
148	Grok 4 Fast	xAI	1412	1401–1423	3K	—
149	Step-3.5-Flash	StepFun	1411	1407–1415	37.5K	$0.10 / $0.30
150	MiniMax-M2.1 Preview	MiniMax	1411	1404–1417	9.3K	$0.30 / $1.20
151	GPT-6 Luna	OpenAI	1410	1405–1415	44.6K	$0.10 / $0.50
153	Grok 4	xAI	1409	1404–1414	19.4K	—
154	Gemini 3.1 Flash-Lite Preview	Google	1407	1403–1412	41.6K	$0.25 / $1.50
155	Grok 4.1 Fast	xAI	1407	1403–1412	33.7K	—
156	Qwen3-VL-235B-A22B-Thinking	Alibaba	1407	1398–1416	4K	—
159	GPT-5 Chat	OpenAI	1404	1399–1410	15.2K	—
161	Qwen3.5-Flash	Alibaba	1403	1399–1408	40.1K	$0.065 / $0.26
162	GPT-4.5 Preview	OpenAI	1403	1393–1413	3.4K	—
163	o3	OpenAI	1402	1397–1407	25.5K	$2 / $8
164	Qwen3.5-35B-A3B	Alibaba	1400	1395–1405	19.3K	$0.14 / $1
165	Solar Pro 4	Upstage	1400	1394–1406	15.8K	$0.15 / $0.60
166	Claude Opus 4	Anthropic	1399	1394–1405	16.2K	—
167	Hunyuan T1	Tencent	1399	1386–1412	2K	—
168	GPT-5.3 Chat	OpenAI	1399	1394–1404	21.6K	—
169	Qwen3-30B-A3B-Instruct	Alibaba	1398	1391–1404	11K	$0.09 / $0.30
170	Muse Glimmer	Meta	1397	1385–1409	2.6K	—
171	Grok 4.3	xAI	1396	1392–1401	47.7K	$1.25 / $2.50
173	Qwen3-235B-A22B	Alibaba	1393	1388–1399	16.4K	$0.09 / $0.55
174	NVIDIA-Nemotron-3-Super-120B-A12B	NVIDIA	1386	1377–1395	4.4K	—
175	GPT-4.1	OpenAI	1384	1379–1389	22K	$2 / $8
176	Kimi K2 Preview	Moonshot AI	1384	1375–1392	5.6K	$0.60 / $2.50
177	GPT-5.4 nano	OpenAI	1381	1376–1385	41.3K	$0.20 / $1.25
178	GPT-5 mini	OpenAI	1380	1374–1386	12.9K	$0.25 / $2
179	GLM-4.5-Air	Z.ai	1379	1373–1384	14.9K	$0.13 / $0.85
180	Gemini 2.5 Flash-Lite Preview	Google	1377	1373–1382	25.5K	$0.10 / $0.40
182	Grok 3 Mini	xAI	1374	1367–1382	7.3K	—
183	Granite 4.2 30B	IBM	1374	1361–1387	2.2K	—
184	Hunyuan TurboS	Tencent	1374	1365–1384	3.9K	—
186	Claude Sonnet 4	Anthropic	1372	1367–1378	15.4K	—
187	MiniMax-M2.5	MiniMax	1372	1367–1377	26.9K	$0.30 / $1.20
188	Qwen3-Coder-480B-A35B-Instruct	Alibaba	1372	1365–1378	11.5K	—
189	Qwen3-Next-80B-A3B-Thinking	Alibaba	1371	1364–1379	6.7K	$0.15 / $1.20
190	o1	OpenAI	1371	1364–1379	6.5K	$15 / $60
191	GLM-4.6V	Z.ai	1368	1353–1382	1.5K	$0.30 / $0.90
193	o3-mini	OpenAI	1365	1356–1374	4.4K	$1.10 / $4.40
194	DeepSeek-V3	DeepSeek	1365	1360–1370	18.6K	$0.29 / $1.14
195	Mistral Medium 3	Mistral AI	1364	1358–1370	13.6K	$0.40 / $2
196	Amazon Nova 2 Lite	Amazon	1364	1356–1372	6.6K	—
197	Grok 3 Mini Beta	xAI	1364	1357–1370	9.2K	—
198	gpt-oss-120b	OpenAI	1363	1358–1369	14.7K	$0.15 / $0.60
199	Ling-flash-2.0	Ant Group	1363	1353–1373	3.5K	—
201	Mercury 2	Inception	1361	1348–1375	1.8K	$0.25 / $0.75
203	Qwen2.5-Max	Alibaba	1359	1353–1366	9.7K	—
204	MiniMax-M2	MiniMax	1357	1347–1367	3.7K	$0.26 / $1.02
205	GLM-4.7-Flash	Z.ai	1356	1349–1363	6.7K	$0.06 / $0.40
208	Step-3	StepFun	1355	1344–1365	3K	—
210	INTELLECT-3	—	1352	1341–1363	2.8K	—
211	o4-mini	OpenAI	1350	1345–1356	19.4K	$1.10 / $4.40
212	Trinity Large Preview	Arcee	1350	1344–1355	19.5K	—
213	GPT-4.1 mini	OpenAI	1349	1343–1355	16.1K	$0.40 / $1.60
214	NVIDIA-Nemotron-3-Nano-30B-A3B	NVIDIA	1348	1341–1355	8.4K	—
215	Gemini 2.0 Flash	Google	1346	1341–1352	14.1K	—
216	Ring-flash-2.0	Ant Group	1344	1334–1354	3.5K	—
217	Trinity Large Thinking	Arcee	1342	1336–1347	20K	$0.25 / $0.80
218	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVIDIA	1341	1335–1348	14.3K	—
219	Gemma 3 27B IT	Google	1340	1335–1345	17.7K	$0.08 / $0.35
220	MiniMax-M1	MiniMax	1339	1334–1345	15.6K	$0.55 / $2.20
221	Llama 3.3 Nemotron Super 49B v1.5	NVIDIA	1335	1321–1350	1.4K	—
222	Mistral Small 3.2	Mistral AI	1335	1328–1342	7.8K	—
223	Step-1o Turbo	StepFun	1335	1325–1345	3.7K	—
224	Qwen3-32B	Alibaba	1334	1318–1350	1.2K	$0.08 / $0.28
225	GLM-4.5V	Z.ai	1334	1322–1346	2.3K	$0.60 / $1.80
226	o1-mini	OpenAI	1333	1327–1339	13.9K	—
227	Claude 3.7 Sonnet	Anthropic	1333	1327–1339	13.7K	—
229	GPT-5 nano	OpenAI	1328	1319–1338	3.8K	$0.05 / $0.40
230	Granite 4.2 8B	IBM	1327	1313–1341	2.1K	$0.06 / $0.25
231	Command A	Cohere	1326	1322–1331	23.8K	$0.30 / $1.50
232	QwQ-32B	Alibaba	1325	1319–1332	8.8K	—
234	Gemini 2.0 Flash-Lite Preview	Google	1324	1317–1332	6.2K	—
235	Olmo 3.1 32B Instruct	Ai2	1322	1314–1330	6.3K	—
236	Qwen-Plus	Alibaba	1316	1302–1331	1.5K	$0.26 / $0.78
237	Llama 3.1 Nemotron Ultra 253B v1	NVIDIA	1316	1295–1337	713	—
239	Qwen3-30B-A3B	Alibaba	1314	1308–1320	10.6K	$0.12 / $0.50
241	Llama 3.3 Nemotron Super 49B v1	NVIDIA	1311	1287–1335	510	—
242	Gemma 3 12B IT	Google	1309	1292–1327	977	$0.05 / $0.15
243	Hunyuan Turbo	Tencent	1308	1285–1331	496	—
244	Granite 4.2 3B	IBM	1307	1292–1321	2K	—
245	Claude 3.5 Sonnet	Anthropic	1305	1301–1309	27.1K	—
246	Yi Lightning	01.AI	1302	1294–1310	7K	—
247	Olmo 3 32B Think	Ai2	1302	1290–1313	3K	—
248	Qwen2.5-Plus	Alibaba	1299	1289–1310	2.7K	—
249	Step-2 16K Exp	StepFun	1299	1284–1314	1.2K	—
250	Gemini 1.5 Pro	Google	1296	1291–1302	14.9K	—
251	GLM-4-Plus	Z.ai	1293	1279–1308	1.5K	—
252	Granite 4.1 8B	IBM	1293	1281–1306	2.7K	—
253	Athene V2 Chat	—	1292	1285–1300	6.5K	—
254	DeepSeek-V2.5	DeepSeek	1289	1276–1302	1.8K	—
255	Molmo 2 8B	Ai2	1287	1260–1314	451	—
256	Hunyuan Large	Tencent	1286	1268–1305	889	—
257	GPT-4.1 nano	OpenAI	1286	1272–1299	1.7K	$0.10 / $0.40
258	Mercury	Inception	1285	1267–1304	1.1K	—
259	Gemma 3n E4B IT	Google	1284	1277–1291	8.5K	—
260	GPT-4o	OpenAI	1281	1275–1286	32.4K	$2.50 / $10
261	Llama 4 Maverick 17B 128E Instruct	Meta	1281	1275–1286	15.9K	—
262	Mistral Small 3.1 24B Instruct	Mistral AI	1278	1272–1284	14.6K	$0.35 / $0.56
264	gpt-oss-20b	OpenAI	1274	1265–1283	4.7K	$0.018 / $0.09
266	Olmo 3.1 32B Think	Ai2	1272	1263–1282	4.4K	—
267	Grok 2	xAI	1272	1266–1277	17.3K	—
268	Qwen2.5-72B-Instruct	Alibaba	1270	1264–1277	10.5K	$0.36 / $0.40
270	Qwen-Max	Alibaba	1269	1260–1278	4.4K	—
271	Llama 3.1 405B Instruct	Meta	1269	1263–1275	10.7K	—
272	GPT-4o mini	OpenAI	1267	1262–1272	18.3K	$0.15 / $0.60
273	Magistral Medium	Mistral AI	1267	1258–1275	5.6K	—
274	Llama 3.1 Nemotron 70B Instruct	NVIDIA	1266	1254–1279	2K	—
275	Llama 4 Scout 17B 16E Instruct	Meta	1266	1260–1272	12.9K	—
276	Hunyuan Standard	Tencent	1264	1246–1283	873	—
279	Gemini 1.5 Flash	Google	1257	1250–1264	9.3K	—
280	Llama 3.3 70B Instruct	Meta	1257	1252–1262	17K	$0.10 / $0.32
281	Mistral Large	Mistral AI	1257	1250–1264	7K	$0.50 / $1.50
282	Hunyuan Large Vision	Tencent	1256	1243–1270	2.2K	—
283	Mistral Large 2	Mistral AI	1256	1250–1263	12.7K	—
284	Grok 2 mini	xAI	1255	1249–1261	14.2K	—
286	Gemma 3 4B IT	Google	1253	1235–1270	1.1K	$0.04 / $0.05
287	Qwen2.5-Coder-32B-Instruct	Alibaba	1251	1237–1265	1.4K	$0.66 / $1
288	Claude 3.5 Haiku	Anthropic	1251	1246–1255	23.5K	—
289	GPT-4 Turbo	OpenAI	1251	1245–1257	28.3K	$10 / $30
290	Gemini Advanced	Google	1248	1241–1255	14.1K	—
291	Amazon Nova Pro 1.0	Amazon	1245	1238–1253	6.3K	—
292	Claude 3 Opus	Anthropic	1245	1241–1250	55.1K	—
293	Athene 70B	—	1243	1234–1252	5.4K	—
295	Llama 3.1 70B Instruct	Meta	1241	1235–1247	15K	$0.40 / $0.40
296	Granite-4.0-H-Small	IBM	1241	1230–1252	3K	—
298	Mistral Small 24B Instruct	Mistral AI	1233	1223–1243	3.6K	$0.05 / $0.08
299	Llama 3.1 Tulu 3 70B	Ai2	1220	1201–1238	779	—
300	Amazon Nova Lite 1.0	Amazon	1220	1211–1228	4.9K	—
302	Phi-4	Microsoft	1219	1212–1227	5.7K	$0.07 / $0.14
303	Hunyuan Standard 256K	Tencent	1218	1198–1238	700	—
304	Jamba 1.5 Large	AI21 Labs	1218	1206–1229	2.4K	—
305	GLM-4	Z.ai	1212	1201–1223	3K	—
306	Reka Core	Reka	1211	1199–1222	2.1K	—
307	Gemini 1.5 Flash-8B	Google	1209	1203–1216	9.7K	—
308	OLMo 2 32B Instruct	Ai2	1208	1188–1227	789	—
309	DeepSeek-Coder-V2	DeepSeek	1207	1197–1216	4.4K	—
310	Llama 3.1 Nemotron 51B Instruct	NVIDIA	1203	1186–1220	1K	—
311	Nemotron-4 340B Instruct	NVIDIA	1201	1193–1210	5.6K	—
312	GPT-4	OpenAI	1200	1192–1208	13.9K	$30 / $60
313	Gemma 2 27B IT	Google	1198	1193–1203	20.5K	$0.65 / $0.65
314	Claude 3 Sonnet	Anthropic	1197	1191–1203	30.7K	—
315	Gemma 2 9B IT SimPO	Google	1196	1185–1207	2.6K	—
316	Llama 3 70B Instruct	Meta	1195	1189–1200	46K	—
317	Aya Expanse 32B	Cohere	1193	1185–1200	7.2K	—
318	Ministral 8B	Mistral AI	1191	1176–1206	1.3K	$0.15 / $0.15
319	Amazon Nova Micro 1.0	Amazon	1191	1183–1199	4.9K	—
320	Qwen2-72B-Instruct	Alibaba	1190	1183–1198	10.8K	—
321	Command R+	Cohere	1185	1175–1196	2.8K	$0.15 / $0.60
322	Reka Flash	Reka	1181	1170–1192	2.3K	—
323	Llama 3.1 8B Instruct	Meta	1175	1168–1181	13.5K	$0.05 / $0.08
325	Llama 3.1 Tulu 3 8B	Ai2	1174	1155–1194	728	—
326	Claude 3 Haiku	Anthropic	1174	1168–1179	33.7K	—
327	Gemma 2 9B IT	Google	1171	1165–1177	14.8K	—
328	Qwen1.5-110B-Chat	Alibaba	1168	1159–1176	7.7K	—
330	Jamba 1.5 Mini	AI21 Labs	1167	1155–1178	2.4K	—
334	Yi 1.5 34B Chat	01.AI	1160	1151–1168	6.7K	—
335	InternLM2.5 20B Chat	—	1158	1147–1169	2.6K	—
336	Aya Expanse 8B	Cohere	1156	1145–1167	2.6K	—
337	Mixtral 8x22B Instruct v0.1	Mistral AI	1150	1143–1157	14.7K	—
338	Mistral Medium	Mistral AI	1149	1141–1158	8.7K	—
339	Qwen1.5-72B-Chat	Alibaba	1148	1140–1156	10.9K	—
340	Granite 3.1 8B Instruct	IBM	1145	1126–1165	798	—
341	Granite 3.1 2B Instruct	IBM	1138	1119–1157	832	—
342	Reka Flash 21B	Reka	1137	1126–1147	4.5K	—
344	Llama 3 8B Instruct	Meta	1133	1127–1139	30.2K	—
345	Qwen1.5-32B-Chat	Alibaba	1130	1121–1140	6.1K	—
346	Phi-3 Medium 4K Instruct	Microsoft	1126	1118–1134	6.8K	—
347	Starling LM 7B Beta	—	1118	1108–1129	4.5K	—
348	Zephyr ORPO 141B A35B v0.1	—	1116	1099–1132	1.2K	—
350	Mixtral 8x7B Instruct v0.1	Mistral AI	1115	1108–1121	19.5K	—
351	Qwen1.5-14B-Chat	Alibaba	1113	1103–1124	4.9K	—
352	DBRX Instruct Preview	—	1113	1104–1122	8.9K	—
353	Gemma 2 2B IT	Google	1112	1106–1119	12.3K	—
354	Gemini Pro	Google	1109	1098–1120	4.5K	$2 / $10
355	GPT-3.5 Turbo	OpenAI	1108	1101–1114	18.5K	$1 / $2
356	Tulu 2 DPO 70B	—	1104	1088–1120	1.4K	—
357	Yi 34B Chat	01.AI	1103	1093–1114	3.8K	—
358	Phi-3 Small 8K Instruct	Microsoft	1100	1091–1110	5.2K	—
360	Llama 3.2 3B Instruct	Meta	1095	1082–1107	2.2K	$0.051 / $0.34
362	Granite 3.0 8B Instruct	IBM	1092	1077–1106	1.7K	—
363	Phi-3 Mini 4K Instruct	Microsoft	1087	1077–1098	3.2K	—
364	OpenChat 3.5	—	1085	1074–1096	3.4K	—
365	WizardLM 70B	Microsoft	1079	1064–1094	1.8K	—
366	Starling LM 7B Alpha	—	1075	1063–1088	2.5K	—
367	Granite 3.0 2B Instruct	IBM	1073	1059–1088	1.7K	—
368	Llama 2 70B Chat	Meta	1073	1065–1081	9.7K	—
370	OpenHermes 2.5 Mistral 7B	Mistral AI	1071	1055–1088	1.1K	—
371	Snowflake Arctic Instruct	Snowflake	1071	1062–1080	9.6K	—
372	Gemma 1.1 7B IT	Google	1071	1062–1079	7K	—
373	DeepSeek LLM 67B Chat	DeepSeek	1070	1051–1088	1.1K	—
374	Mistral 7B Instruct v0.2	Mistral AI	1067	1057–1077	5.2K	—
375	Vicuna 33B	—	1067	1057–1077	5.1K	—
377	Qwen1.5-7B-Chat	Alibaba	1065	1050–1081	1.3K	—
378	SOLAR 10.7B Instruct v1.0	Upstage	1062	1043–1081	932	—
379	Dolphin 2.2.1 Mistral 7B	Mistral AI	1054	1027–1081	359	—
380	Nous Hermes 2 Mixtral 8x7B DPO	Nous Research	1053	1036–1071	1.1K	—
381	SmolLM2 1.7B Instruct	—	1053	1029–1077	572	—
382	Code Llama 70B Instruct	Meta	1052	1022–1083	306	—
383	MPT 30B Chat	—	1051	1025–1078	434	—
384	Llama 2 13B Chat	Meta	1051	1041–1061	4.5K	—
385	Llama 2 70B SteerLM Chat	NVIDIA	1047	1026–1067	787	—
386	Llama 3.2 1B Instruct	Meta	1044	1032–1057	2.2K	$0.027 / $0.20
387	Gemma 7B IT	Google	1042	1028–1055	2.3K	—
388	Code Llama 34B Instruct	Meta	1032	1017–1046	1.5K	—
389	Phi-3 Mini 128K Instruct	Microsoft	1028	1017–1039	6.2K	—
390	Qwen-14B-Chat	Alibaba	1027	1008–1046	1K	—
391	Zephyr 7B Beta	—	1026	1012–1041	2.2K	—
392	Zephyr 7B Alpha	—	1024	995–1052	333	—
393	Vicuna 13B	—	1023	1013–1034	4K	—
394	WizardLM 13B	Microsoft	1018	1001–1034	1.3K	—
395	Llama 2 7B Chat	Meta	1009	998–1020	3.3K	—
396	Falcon 180B Chat	TII	1007	974–1041	231	—
397	Guanaco 33B	—	1005	980–1029	517	—
398	Gemma 1.1 2B IT	Google	1005	993–1017	3.2K	—
399	PaLM 2	Google	1005	989–1020	1.6K	—
400	Mistral 7B Instruct	Mistral AI	1001	986–1017	1.9K	—
401	StripedHyena Nous 7B	—	998	981–1015	1.2K	—
402	Vicuna 7B	—	995	978–1011	1.3K	—
403	OLMo 7B Instruct	Ai2	993	976–1009	1.5K	—
404	Gemma 2B IT	Google	989	971–1006	1.2K	—
405	Qwen1.5-4B-Chat	Alibaba	976	962–990	2.1K	—
406	ChatGLM3 6B	—	953	933–973	999	—
407	GPT4All 13B Snoozy	OpenAI	936	907–965	347	—
408	Koala 13B	—	930	913–948	1.3K	—
409	ChatGLM2 6B	—	917	891–943	490	—
410	ChatGLM 6B	—	901	880–922	848	—
411	RWKV-4 Raven 14B	—	896	876–916	876	—
412	MPT 7B Chat	—	895	874–917	714	—
413	OpenAssistant Pythia 12B	—	884	866–903	1.2K	—
414	StableLM Tuned Alpha 7B	—	835	811–859	575	—
415	Alpaca 13B	—	826	806–845	1.1K	—
416	FastChat-T5 3B	—	814	793–835	797	—
417	Dolly 2.0 12B	—	804	780–827	615	—
418	LLaMA 13B	Meta	728	699–757	430	—

LMArena ranks overlap where confidence intervals do, so several models can share a rank. Where a model is rated at several effort levels, its best is shown.

Full leaderboard

Related leaderboards