LMArena Creative writing leaderboard

LMArena’s text votes on stories, poems and other creative writing. Claude Opus 5.5 leads with a rating of 1534.

Models: 349
Leader: Claude Opus 5.5
Source: LMArena
As of: Oct 22, 2026

LMArena Creative writing, 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	Claude Opus 5.5	Anthropic	1534	1521–1548	2.5K	$4 / $20
2	Gemini 4 Argon	Google	1523	1510–1537	2.3K	$2 / $10
3	Claude Fable 5.1	Anthropic	1510	1499–1521	3.7K	$10 / $50
4	Claude Opus 4.6	Anthropic	1505	1499–1511	15.4K	$5 / $25
5	Claude Opus 5	Anthropic	1495	1487–1503	7.9K	$5 / $25
6	Claude Fable 5	Anthropic	1493	1486–1501	9.6K	$10 / $50
7	Muse Wonder	Meta	1493	1476–1510	1.4K	$4 / $20
8	Gemini 3.8 Flash	Google	1489	1482–1497	9.8K	$0.82 / $4.13
9	Gemini 3.7 Flash	Google	1489	1481–1498	6.8K	$0.75 / $3.75
11	Claude Opus 4.7	Anthropic	1486	1479–1493	11.8K	$5 / $25
13	Gemini 3 Pro	Google	1483	1474–1491	6.5K	—
14	Gemini 3.1 Pro Preview	Google	1482	1477–1487	24.4K	$2 / $12
16	Qwen3.8-Max	Alibaba	1477	1469–1486	6.1K	$2 / $6
17	Gemini 3.5 Flash	Google	1470	1463–1477	9.5K	$1.50 / $9
18	Claude Sonnet 5.5	Anthropic	1468	1454–1482	2.1K	$2 / $10
20	MiMo-V2.6-Pro	Xiaomi	1466	1445–1488	792	$0.43 / $0.87
21	Gemini 3.6 Flash	Google	1466	1458–1474	7.6K	$0.75 / $3.75
22	Qwen3.5-Max-Preview	Alibaba	1464	1454–1475	3.3K	—
23	Muse Spark 1.3	Meta	1462	1453–1472	4.6K	$1.25 / $4.25
24	GLM-5.2	Z.ai	1462	1455–1469	10.5K	$0.034 / $7
25	GLM-5.3	Z.ai	1459	1450–1468	5.7K	$0.079 / $1.40
26	Muse Spark	Meta	1459	1445–1472	2K	—
27	Gemini 3 Flash	Google	1457	1448–1466	4.8K	$0.50 / $3
28	Kimi K3	Moonshot AI	1456	1448–1464	7.2K	$0.24 / $11
29	GPT-5.5	OpenAI	1455	1449–1462	13.2K	$5 / $30
30	Claude Opus 4.8	Anthropic	1454	1448–1461	12.8K	$5 / $25
31	Gemini 2.5 Pro	Google	1454	1449–1459	17.7K	$1.25 / $10
32	GLM-5.1	Z.ai	1453	1446–1459	12.8K	$1.40 / $4.40
34	GPT-5.6 Sol	OpenAI	1450	1443–1458	9K	$4 / $20
35	Qwen3.7-Max-Preview	Alibaba	1448	1421–1474	511	$1.48 / $4.42
37	DeepSeek-V4-Pro	DeepSeek	1446	1439–1453	9.7K	$0.66 / $1.98
38	Claude Opus 4.5	Anthropic	1445	1437–1453	5.7K	$5 / $25
40	Grok 4.5	xAI	1442	1434–1449	8.3K	$2 / $6
41	Claude Sonnet 4.5	Anthropic	1442	1436–1448	12.4K	$3 / $15
42	ERNIE 5.1	Baidu	1441	1433–1449	6.5K	—
44	GLM-5.3-Flash	Z.ai	1441	1433–1449	8.2K	$0.15 / $0.50
45	GPT-5.4	OpenAI	1439	1432–1446	11.1K	$2.50 / $15
47	DeepSeek-V4.1-Flash-20260910	DeepSeek	1439	1427–1450	2.8K	$0.15 / $0.60
48	Grok 4.20 Beta 1	xAI	1438	1429–1448	4.2K	—
49	Qwen3.7-Plus	Alibaba	1438	1430–1446	7.8K	$0.32 / $1.28
50	MiMo-V2.5-Pro	Xiaomi	1438	1432–1444	14.6K	$0.30 / $0.61
51	GLM-5	Z.ai	1438	1429–1447	4.9K	$1 / $3.20
52	Grok 4.20 Multi-Agent Beta	xAI	1436	1430–1443	11.3K	—
53	Claude Sonnet 4.6	Anthropic	1435	1429–1442	12.4K	$3 / $15
54	Qwen3.6-Max-Preview	Alibaba	1435	1413–1458	757	—
56	Muse Spark 1.2	Meta	1435	1421–1448	2.2K	$1.25 / $4.25
57	Muse Spark 1.1	Meta	1435	1428–1442	9.7K	$1.25 / $4.25
58	Kimi K2.6	Moonshot AI	1434	1425–1442	6.3K	$0.65 / $3.41
59	Grok 4.20 Beta	xAI	1432	1425–1439	11.4K	—
60	Grok 4.6	xAI	1428	1420–1436	7.2K	$2 / $6
62	GPT-5.1	OpenAI	1427	1419–1435	6.2K	$1.25 / $10
63	ERNIE 5.0	Baidu	1426	1418–1434	5.8K	—
65	GPT-6.1 Sol	OpenAI	1424	1410–1439	1.9K	$2 / $10
67	Kimi K2.5	Moonshot AI	1422	1416–1428	12.3K	$0.50 / $2.50
68	Gemini 3.5 Flash-Lite	Google	1421	1413–1429	7.5K	$0.30 / $2.50
69	GPT-6 Astra	OpenAI	1421	1409–1432	3.2K	$10 / $50
70	GPT-5.5 Instant	OpenAI	1418	1409–1428	4.4K	—
71	GLM-5V-Turbo	Z.ai	1418	1404–1432	2K	$1.20 / $4
73	Claude Sonnet 5	Anthropic	1416	1409–1423	10.8K	$2 / $10
74	Gemma 4 31B	Google	1416	1397–1435	971	$0.20 / $0.40
75	MiMo-V2-Pro	Xiaomi	1415	1405–1425	3.7K	—
76	Grok 3 Preview	xAI	1414	1405–1423	4.7K	—
77	GPT-5.6 Terra	OpenAI	1413	1406–1420	9.9K	$2 / $12
78	Claude Opus 4.1	Anthropic	1412	1404–1420	6.9K	$15 / $75
79	GLM-4.6	Z.ai	1412	1403–1420	5.1K	$0.55 / $2.20
80	Step 5 Preview	StepFun	1411	1398–1425	2.1K	$1 / $2.70
81	Grok 4.1	xAI	1411	1404–1418	10.1K	—
83	Dola Seed 2.0 Pro	ByteDance	1406	1400–1412	13.2K	—
84	DeepSeek-R1	DeepSeek	1405	1393–1418	2.4K	$0.50 / $2.15
85	ChatGPT-4o Latest	OpenAI	1405	1399–1411	11.5K	—
86	Hy3	Tencent	1405	1393–1417	2.7K	$0.083 / $0.33
87	MiMo-V2.6-Flash	Xiaomi	1404	1392–1416	2.7K	$0.14 / $0.28
88	Qwen3.6-Plus	Alibaba	1404	1396–1412	7.2K	$0.33 / $1.95
91	DeepSeek-V3.1-Terminus	DeepSeek	1403	1376–1431	446	$0.27 / $1
92	DeepSeek-V3.2-Exp	DeepSeek	1403	1389–1418	1.7K	$0.28 / $0.42
93	DeepSeek-V4-Flash	DeepSeek	1403	1395–1410	8.6K	$0.042 / $0.18
94	MiniMax-M3	MiniMax	1403	1396–1409	12.4K	$0.30 / $1.20
95	Qwen3.5-397B-A17B	Alibaba	1403	1397–1408	17.2K	$0.50 / $3.60
96	GLM-4.7	Z.ai	1401	1388–1414	2K	$0.54 / $1.98
97	Gemma 4 26B A4B	Google	1401	1383–1419	1K	$0.068 / $0.23
98	Qwen3-Max-Preview	Alibaba	1401	1391–1411	3.7K	—
99	Gemini 3.1 Flash-Lite Preview	Google	1401	1394–1408	10.6K	$0.25 / $1.50
100	GPT-5.2 Chat	OpenAI	1401	1393–1409	5.6K	$1.75 / $14
101	DeepSeek-V3.1	DeepSeek	1401	1386–1416	1.6K	$0.25 / $0.95
102	Gemini 2.5 Flash	Google	1400	1395–1405	17.6K	$0.30 / $2.50
105	NVIDIA-Nemotron-3-Ultra-550B-A55B	NVIDIA	1398	1385–1412	2.2K	—
106	Grok 4	xAI	1397	1388–1405	5.3K	—
107	GPT-5.6 Luna	OpenAI	1396	1388–1403	8.7K	$0.20 / $1.20
108	GLM-4.5	Z.ai	1395	1384–1406	3.1K	$0.60 / $2.20
109	Grok 4.1 Fast	xAI	1394	1388–1401	9K	—
110	GPT-4.5 Preview	OpenAI	1394	1382–1406	2.6K	—
111	Hunyuan T1	Tencent	1393	1369–1417	616	—
112	MiMo-V2.5	Xiaomi	1393	1385–1400	7.9K	$0.12 / $0.24
113	MiMo-V2-Omni	Xiaomi	1392	1382–1403	3.4K	—
114	Beam 1.0 501B A23B	Reflection AI	1391	1371–1411	968	$0.25 / $1
116	Mistral Medium 3.1	Mistral AI	1391	1385–1396	14K	$0.40 / $2
117	Inkling	Thinky	1390	1383–1397	9.4K	$1 / $4.05
118	Hunyuan Vision 1.5 Thinking	Tencent	1389	1353–1426	250	—
121	LongCat-Flash-Chat Exp	Meituan	1388	1379–1397	4.4K	—
123	Mistral Large 3	Mistral AI	1387	1381–1392	15.7K	$0.50 / $1.50
125	Grok 4 Fast	xAI	1386	1366–1406	889	—
126	Claude Opus 4	Anthropic	1386	1377–1395	4.5K	—
127	Qwen3-235B-A22B-Thinking	Alibaba	1384	1366–1401	1.1K	$0.45 / $3.50
128	Qwen3.8-27B	Alibaba	1383	1374–1392	5.3K	$0.20 / $2.50
130	Grok 4.7	xAI	1381	1370–1393	3.3K	$2 / $6
132	Grok 4.3	xAI	1380	1373–1386	13.2K	$1.25 / $2.50
134	GPT-6 Sol	OpenAI	1377	1366–1389	3.1K	$2 / $10
135	MiMo-V2-Flash	Xiaomi	1376	1369–1383	7.3K	—
136	GPT-5.2	OpenAI	1376	1369–1383	7.7K	$1.75 / $14
138	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B	NVIDIA	1376	1356–1396	956	—
139	Qwen3-235B-A22B-Instruct	Alibaba	1375	1369–1380	13.9K	$0.09 / $0.55
141	Mistral Medium 3.5	Mistral AI	1374	1360–1388	2.1K	$1.50 / $7.50
142	Kimi K2 Thinking Turbo	Moonshot AI	1373	1367–1380	9.5K	—
143	Claude Haiku 4.5	Anthropic	1371	1366–1375	26.2K	$1 / $5
144	GPT-5.4 mini	OpenAI	1370	1363–1377	10.8K	$0.75 / $4.50
145	amazon-nova-experimental-chat	Amazon	1369	1343–1394	481	—
146	Qwen3.5-122B-A10B	Alibaba	1368	1359–1377	4.5K	$0.40 / $3.20
147	Gemini 2.5 Flash-Lite Preview	Google	1367	1358–1376	4.2K	$0.10 / $0.40
148	Qwen3-VL-235B-A22B-Instruct	Alibaba	1366	1349–1383	1.4K	$0.21 / $1.90
149	GPT-5 Chat	OpenAI	1365	1355–1374	4K	—
150	GPT-5	OpenAI	1364	1355–1374	4.2K	$1.25 / $10
151	DeepSeek-V3	DeepSeek	1364	1356–1372	6K	$0.29 / $1.14
152	Mistral Large 4	Mistral AI	1364	1351–1377	2.4K	$1.36 / $4.18
153	GPT-4.1	OpenAI	1363	1356–1371	6.7K	$2 / $8
154	GPT-6 Luna	OpenAI	1363	1356–1369	16.4K	$0.10 / $0.50
155	Qwen3.5-27B	Alibaba	1363	1353–1372	4.3K	$0.27 / $2.16
156	MiniMax-M2.1 Preview	MiniMax	1362	1351–1373	2.8K	$0.30 / $1.20
157	o3	OpenAI	1358	1351–1366	7.6K	$2 / $8
158	Hunyuan TurboS	Tencent	1358	1343–1374	1.5K	—
160	Step-3.5-Flash	StepFun	1357	1351–1364	9.8K	$0.10 / $0.30
161	Qwen3-235B-A22B	Alibaba	1356	1347–1365	4.6K	$0.09 / $0.55
162	GPT-5.3 Chat	OpenAI	1355	1347–1364	5.4K	—
163	MiniMax-M2.7	MiniMax	1354	1348–1360	15.1K	$0.30 / $1.20
166	Kimi K2 Preview	Moonshot AI	1350	1334–1365	1.5K	$0.60 / $2.50
169	GLM-4.6V	Z.ai	1347	1319–1376	425	$0.30 / $0.90
170	Qwen3.5-35B-A3B	Alibaba	1347	1338–1356	4.7K	$0.14 / $1
171	o1	OpenAI	1347	1338–1356	4.6K	$15 / $60
173	Gemma 3 27B IT	Google	1345	1338–1353	6.3K	$0.08 / $0.35
174	Claude Sonnet 4	Anthropic	1345	1335–1354	4.3K	—
175	Qwen3-VL-235B-A22B-Thinking	Alibaba	1344	1326–1362	1K	—
176	Mistral Medium 3	Mistral AI	1343	1334–1353	4.1K	$0.40 / $2
177	Qwen3.5-Flash	Alibaba	1343	1336–1350	9.7K	$0.065 / $0.26
178	GLM-4.5-Air	Z.ai	1343	1334–1353	4K	$0.13 / $0.85
180	Grok 3 Mini Beta	xAI	1342	1330–1353	2.6K	—
182	Gemini 2.0 Flash	Google	1339	1332–1347	6.3K	—
183	Qwen2.5-Max	Alibaba	1338	1330–1347	5K	—
184	Muse Glimmer	Meta	1338	1316–1361	734	—
185	Claude Haiku 5.5	Anthropic	1336	1319–1352	1.4K	$0.10 / $0.50
186	Grok 3 Mini	xAI	1334	1321–1348	1.9K	—
187	Inkling-Small	Thinky	1334	1326–1342	7.7K	$0.45 / $1.20
188	Qwen3-Next-80B-A3B-Instruct	Alibaba	1333	1323–1344	3K	$0.09 / $1.10
189	Qwen3-Coder-480B-A35B-Instruct	Alibaba	1333	1322–1344	3.2K	—
190	Gemini 1.5 Pro	Google	1332	1325–1340	8.1K	—
191	Claude 3.7 Sonnet	Anthropic	1332	1323–1340	5.1K	—
192	Gemma 3 12B IT	Google	1331	1308–1354	600	$0.05 / $0.15
193	MiniMax-M2.5	MiniMax	1331	1323–1338	6.7K	$0.30 / $1.20
196	Step-2 16K Exp	StepFun	1327	1307–1348	751	—
197	GPT-5 mini	OpenAI	1325	1315–1335	3.5K	$0.25 / $2
199	Step-3	StepFun	1322	1301–1342	772	—
200	Trinity Large Preview	Arcee	1320	1311–1329	4.8K	—
201	Command A	Cohere	1319	1312–1326	7.6K	$0.30 / $1.50
202	Gemini 2.0 Flash-Lite Preview	Google	1319	1310–1328	4.1K	—
205	Solar Pro 4	Upstage	1318	1308–1327	5.3K	$0.15 / $0.60
206	Qwen3-30B-A3B-Instruct	Alibaba	1317	1306–1328	3K	$0.09 / $0.30
207	NVIDIA-Nemotron-3-Super-120B-A12B	NVIDIA	1317	1299–1334	1.1K	—
210	INTELLECT-3	—	1316	1295–1337	834	—
211	GPT-5.4 nano	OpenAI	1314	1307–1321	10.4K	$0.20 / $1.25
212	Qwen3-Next-80B-A3B-Thinking	Alibaba	1314	1300–1328	1.8K	$0.15 / $1.20
213	Llama 3.1 Nemotron Ultra 253B v1	NVIDIA	1314	1288–1340	444	—
214	GLM-4-Plus	Z.ai	1311	1292–1330	918	—
215	Trinity Large Thinking	Arcee	1308	1299–1318	4.7K	$0.25 / $0.80
216	Step-1o Turbo	StepFun	1307	1288–1326	992	—
217	Llama 3.3 Nemotron Super 49B v1.5	NVIDIA	1307	1278–1336	370	—
219	Mistral Small 3.2	Mistral AI	1305	1292–1317	2.1K	—
220	GPT-4.1 mini	OpenAI	1299	1291–1308	5.1K	$0.40 / $1.60
221	MiniMax-M1	MiniMax	1298	1288–1307	4.5K	$0.55 / $2.20
222	Qwen3-32B	Alibaba	1297	1275–1319	616	$0.08 / $0.28
223	GLM-4.7-Flash	Z.ai	1296	1282–1310	1.9K	$0.06 / $0.40
224	GLM-4.5V	Z.ai	1295	1272–1318	613	$0.60 / $1.80
225	o4-mini	OpenAI	1294	1286–1302	5.8K	$1.10 / $4.40
226	Amazon Nova 2 Lite	Amazon	1292	1278–1307	1.8K	—
227	Qwen-Plus	Alibaba	1292	1274–1311	931	$0.26 / $0.78
228	Claude 3.5 Sonnet	Anthropic	1292	1286–1298	11.7K	—
229	GPT-4o	OpenAI	1291	1284–1298	16.6K	$2.50 / $10
230	Mercury 2	Inception	1289	1263–1314	547	$0.25 / $0.75
231	Llama 3.3 Nemotron Super 49B v1	NVIDIA	1288	1261–1315	428	—
232	Granite 4.2 30B	IBM	1288	1263–1313	661	—
233	QwQ-32B	Alibaba	1287	1278–1297	3.7K	—
234	Ling-flash-2.0	Ant Group	1287	1267–1307	854	—
235	Gemma 3n E4B IT	Google	1287	1276–1298	2.9K	—
236	MiniMax-M2	MiniMax	1287	1267–1307	895	$0.26 / $1.02
237	o3-mini	OpenAI	1286	1275–1296	3.1K	$1.10 / $4.40
238	Gemini 1.5 Flash	Google	1285	1276–1294	5K	—
239	DeepSeek-V2.5	DeepSeek	1285	1268–1302	1.1K	—
240	Grok 2	xAI	1284	1277–1291	9.3K	—
241	Gemini Advanced	Google	1284	1274–1294	7K	—
242	Yi Lightning	01.AI	1280	1270–1290	3.8K	—
243	gpt-oss-120b	OpenAI	1274	1265–1284	3.9K	$0.15 / $0.60
247	Qwen3-30B-A3B	Alibaba	1271	1261–1282	3.3K	$0.12 / $0.50
248	Gemma 3 4B IT	Google	1270	1248–1292	666	$0.04 / $0.05
250	Ring-flash-2.0	Ant Group	1270	1250–1290	888	—
251	NVIDIA-Nemotron-3-Nano-30B-A3B	NVIDIA	1270	1257–1282	2.4K	—
252	Hunyuan Turbo	Tencent	1269	1244–1294	451	—
253	Llama 3.1 Nemotron 70B Instruct	NVIDIA	1268	1250–1287	1K	—
254	GPT-4 Turbo	OpenAI	1268	1261–1276	14.4K	$10 / $30
255	GPT-4o mini	OpenAI	1268	1261–1275	10.5K	$0.15 / $0.60
256	Llama 4 Maverick 17B 128E Instruct	Meta	1266	1258–1275	5.2K	—
257	Olmo 3.1 32B Instruct	Ai2	1265	1250–1279	1.7K	—
258	Hunyuan Large	Tencent	1264	1241–1288	568	—
260	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVIDIA	1263	1253–1273	4.9K	—
261	Llama 3.1 405B Instruct	Meta	1262	1254–1269	8.9K	—
262	Qwen2.5-Plus	Alibaba	1262	1247–1276	1.6K	—
263	GPT-4.1 nano	OpenAI	1260	1242–1278	1K	$0.10 / $0.40
265	Olmo 3 32B Think	Ai2	1256	1235–1277	863	—
266	Mistral Small 3.1 24B Instruct	Mistral AI	1252	1243–1262	4.1K	$0.35 / $0.56
267	Granite 4.1 8B	IBM	1250	1226–1275	696	—
268	Llama 3.3 70B Instruct	Meta	1250	1243–1257	8K	$0.10 / $0.32
269	GPT-5 nano	OpenAI	1250	1230–1269	910	$0.05 / $0.40
270	Llama 4 Scout 17B 16E Instruct	Meta	1249	1239–1259	3.8K	—
271	Qwen-Max	Alibaba	1248	1236–1260	2.4K	—
272	Hunyuan Large Vision	Tencent	1248	1224–1272	628	—
273	Magistral Medium	Mistral AI	1245	1229–1262	1.4K	—
275	o1-mini	OpenAI	1244	1236–1252	7.9K	—
276	Mistral Large 2	Mistral AI	1243	1234–1251	6.7K	—
277	Hunyuan Standard	Tencent	1242	1219–1265	590	—
278	Mistral Large	Mistral AI	1242	1233–1251	4.4K	$0.50 / $1.50
279	Grok 2 mini	xAI	1241	1234–1249	7.9K	—
280	Gemma 2 27B IT	Google	1240	1234–1247	11.4K	$0.65 / $0.65
281	Gemma 2 9B IT SimPO	Google	1239	1224–1253	1.7K	—
282	Athene 70B	—	1238	1227–1249	3.1K	—
285	Reka Core	Reka	1236	1218–1254	1K	—
286	Granite 4.2 8B	IBM	1235	1208–1262	635	$0.06 / $0.25
287	Claude 3 Opus	Anthropic	1235	1228–1241	28.4K	—
288	Command R+	Cohere	1234	1219–1249	1.5K	$0.15 / $0.60
289	Athene V2 Chat	—	1234	1224–1244	3.7K	—
290	Claude 3.5 Haiku	Anthropic	1232	1226–1239	9.8K	—
292	Llama 3.1 70B Instruct	Meta	1232	1224–1239	8.3K	$0.40 / $0.40
293	Llama 3.1 Tulu 3 70B	Ai2	1231	1207–1255	441	—
294	Olmo 3.1 32B Think	Ai2	1227	1210–1244	1.4K	—
296	Qwen2.5-72B-Instruct	Alibaba	1221	1213–1229	5.7K	$0.36 / $0.40
297	Gemini 1.5 Flash-8B	Google	1217	1208–1226	5K	—
298	Llama 3.1 Nemotron 51B Instruct	NVIDIA	1213	1188–1238	482	—
299	Amazon Nova Pro 1.0	Amazon	1212	1203–1222	3.9K	—
300	Jamba 1.5 Large	AI21 Labs	1212	1196–1227	1.4K	—
301	Granite-4.0-H-Small	IBM	1210	1186–1233	706	—
302	Llama 3 70B Instruct	Meta	1210	1202–1217	22.8K	—
303	Granite 4.2 3B	IBM	1207	1179–1235	637	—
304	Gemma 2 9B IT	Google	1205	1198–1213	8.3K	—
305	Nemotron-4 340B Instruct	NVIDIA	1202	1191–1214	3.2K	—
306	Reka Flash	Reka	1202	1185–1219	1K	—
307	gpt-oss-20b	OpenAI	1201	1183–1219	1.2K	$0.018 / $0.09
309	Aya Expanse 32B	Cohere	1200	1190–1210	3.9K	—
310	OLMo 2 32B Instruct	Ai2	1198	1174–1223	567	—
311	GLM-4	Z.ai	1198	1183–1213	1.5K	—
312	Amazon Nova Lite 1.0	Amazon	1197	1186–1207	3K	—
313	Mistral Small 24B Instruct	Mistral AI	1194	1182–1207	2.4K	$0.05 / $0.08
314	GPT-4	OpenAI	1191	1183–1200	13.9K	$30 / $60
316	Mercury	Inception	1189	1151–1227	295	—
317	Claude 3 Sonnet	Anthropic	1186	1178–1194	15.6K	—
318	Phi-4	Microsoft	1181	1172–1191	4.1K	$0.07 / $0.14
319	Llama 3.1 Tulu 3 8B	Ai2	1181	1156–1206	472	—
320	Qwen2-72B-Instruct	Alibaba	1181	1171–1190	5.5K	—
321	Ministral 8B	Mistral AI	1175	1153–1196	678	$0.15 / $0.15
322	Qwen2.5-Coder-32B-Instruct	Alibaba	1174	1155–1193	836	$0.66 / $1
323	Amazon Nova Micro 1.0	Amazon	1172	1161–1182	3.1K	—
325	Aya Expanse 8B	Cohere	1168	1153–1183	1.5K	—
326	Hunyuan Standard 256K	Tencent	1165	1136–1193	357	—
327	Jamba 1.5 Mini	AI21 Labs	1164	1149–1180	1.4K	—
329	Mistral Medium	Mistral AI	1158	1147–1168	5.5K	—
330	Claude 3 Haiku	Anthropic	1157	1150–1165	16.7K	—
331	Llama 3.1 8B Instruct	Meta	1154	1146–1162	7.3K	$0.05 / $0.08
333	Llama 3 8B Instruct	Meta	1150	1142–1158	15.4K	—
334	WizardLM 70B	Microsoft	1149	1132–1166	1.8K	—
335	Qwen1.5-110B-Chat	Alibaba	1148	1136–1159	3.7K	—
336	Gemma 2 2B IT	Google	1147	1138–1155	7.1K	—
337	Mixtral 8x22B Instruct v0.1	Mistral AI	1141	1132–1150	7.4K	—
338	Qwen1.5-72B-Chat	Alibaba	1137	1127–1147	5.9K	—
339	Yi 1.5 34B Chat	01.AI	1135	1124–1146	3.6K	—
341	Reka Flash 21B	Reka	1130	1116–1145	2.1K	—
342	Gemini Pro	Google	1130	1116–1144	3K	$2 / $10
343	Granite 3.1 8B Instruct	IBM	1129	1102–1155	450	—
344	OpenChat 3.5	—	1125	1107–1143	1.5K	—
345	Vicuna 33B	—	1124	1113–1136	4K	—
347	InternLM2.5 20B Chat	—	1123	1107–1138	1.4K	—
348	DeepSeek-Coder-V2	DeepSeek	1120	1107–1133	2.3K	—
349	Granite 3.1 2B Instruct	IBM	1116	1087–1145	442	—
350	Zephyr ORPO 141B A35B v0.1	—	1113	1090–1136	635	—
351	Mixtral 8x7B Instruct v0.1	Mistral AI	1109	1100–1117	11.2K	—
352	Yi 34B Chat	01.AI	1108	1094–1122	2.4K	—
353	Phi-3 Medium 4K Instruct	Microsoft	1106	1096–1117	3.7K	—
354	Tulu 2 DPO 70B	—	1105	1086–1124	1.1K	—
355	DBRX Instruct Preview	—	1104	1092–1116	4.8K	—
357	Zephyr 7B Beta	—	1102	1086–1118	2K	—
358	Nous Hermes 2 Mixtral 8x7B DPO	Nous Research	1101	1078–1125	616	—
359	Starling LM 7B Alpha	—	1099	1084–1114	1.7K	—
360	OpenHermes 2.5 Mistral 7B	Mistral AI	1097	1075–1119	811	—
361	Starling LM 7B Beta	—	1095	1081–1110	2.4K	—
362	Llama 3.2 3B Instruct	Meta	1094	1075–1113	1.1K	$0.051 / $0.34
364	GPT-3.5 Turbo	OpenAI	1092	1083–1101	9.7K	$1 / $2
365	Qwen1.5-14B-Chat	Alibaba	1090	1076–1104	2.4K	—
366	Llama 2 70B SteerLM Chat	NVIDIA	1090	1066–1115	604	—
367	WizardLM 13B	Microsoft	1090	1072–1108	1.4K	—
368	SOLAR 10.7B Instruct v1.0	Upstage	1089	1065–1113	698	—
369	Falcon 180B Chat	TII	1088	1052–1125	270	—
370	Dolphin 2.2.1 Mistral 7B	Mistral AI	1086	1052–1119	291	—
371	Qwen1.5-32B-Chat	Alibaba	1083	1070–1095	3K	—
372	Phi-3 Small 8K Instruct	Microsoft	1083	1070–1096	2.4K	—
373	Snowflake Arctic Instruct	Snowflake	1081	1069–1093	4.8K	—
374	Guanaco 33B	—	1080	1053–1107	515	—
375	Llama 2 70B Chat	Meta	1075	1065–1085	6.4K	—
376	MPT 30B Chat	—	1074	1045–1102	452	—
377	Granite 3.0 8B Instruct	IBM	1070	1050–1091	895	—
378	Zephyr 7B Alpha	—	1069	1039–1098	399	—
379	Mistral 7B Instruct v0.2	Mistral AI	1067	1055–1080	3K	—
380	DeepSeek LLM 67B Chat	DeepSeek	1067	1045–1089	789	—
381	Gemma 1.1 7B IT	Google	1059	1048–1071	3.4K	—
382	Vicuna 13B	—	1058	1045–1070	3.3K	—
383	Llama 2 13B Chat	Meta	1047	1034–1059	3.3K	—
384	Granite 3.0 2B Instruct	IBM	1046	1025–1067	916	—
385	Phi-3 Mini 4K Instruct	Microsoft	1044	1030–1059	2K	—
388	Qwen1.5-7B-Chat	Alibaba	1034	1012–1057	701	—
389	Llama 2 7B Chat	Meta	1033	1019–1047	2.3K	—
390	Llama 3.2 1B Instruct	Meta	1033	1013–1052	1.1K	$0.027 / $0.20
391	Code Llama 34B Instruct	Meta	1031	1015–1048	1.5K	—
392	SmolLM2 1.7B Instruct	—	1031	997–1065	341	—
393	StripedHyena Nous 7B	—	1029	1008–1050	869	—
394	Mistral 7B Instruct	Mistral AI	1028	1011–1045	1.7K	—
395	Qwen-14B-Chat	Alibaba	1027	1007–1048	898	—
396	Phi-3 Mini 128K Instruct	Microsoft	1024	1010–1038	2.9K	—
397	Gemma 7B IT	Google	1023	1006–1041	1.3K	—
398	Vicuna 7B	—	1021	1003–1039	1.3K	—
399	Gemma 1.1 2B IT	Google	998	981–1015	1.5K	—
400	PaLM 2	Google	991	974–1009	1.6K	—
401	OLMo 7B Instruct	Ai2	990	969–1010	1K	—
402	Gemma 2B IT	Google	986	963–1009	696	—
403	Koala 13B	—	973	953–993	1.2K	—
404	GPT4All 13B Snoozy	OpenAI	972	934–1009	250	—
405	ChatGLM3 6B	—	970	947–993	816	—
406	Qwen1.5-4B-Chat	Alibaba	969	950–988	1.1K	—
407	Alpaca 13B	—	967	945–989	954	—
408	MPT 7B Chat	—	957	934–981	683	—
409	ChatGLM2 6B	—	941	915–967	566	—
410	RWKV-4 Raven 14B	—	934	911–956	874	—
411	OpenAssistant Pythia 12B	—	923	903–943	1.1K	—
412	FastChat-T5 3B	—	902	879–925	744	—
413	ChatGLM 6B	—	900	876–924	771	—
414	StableLM Tuned Alpha 7B	—	868	840–895	536	—
415	Dolly 2.0 12B	—	864	837–890	565	—
416	LLaMA 13B	Meta	794	762–826	431	—

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