所有650 LLMs从智能和成本上进行比较

来自 Artificial Analysis 和其他来源的 2026-09-22 快照中的每个模型,按照运行基准测试所花费的每美元的智能程度进行排名。带星号的行包含智能与成本 Pareto frontier。打开一个模型,看看它的数字,是否仍然是最佳价值,以及正面对决。“开放权重前沿”标记的是没有其他可下载权重模型能超越的可下载权重模型。

模型
650
可下载权重的模型 (open weights)
364
在帕累托前沿
14
可下载权重模型前沿(open-weights frontier)
6
数据快照
2026-09-22
今天的Pareto frontier

各个预算下的最佳性价比

每个模型都是同等价格下最智能的选择,其他所有模型都至少被其中一个超越。按价格从低到高排列。

  1. IBM可下载权重的模型 (open weights)Granite 4.2 3B
    智能得分 (Intelligence Index)
    9.10
    基准测试费用(美元)
    $14.9
  2. OpenAIGPT-5.6 Luna (Non-reasoning)
    智能得分 (Intelligence Index)
    15.50
    基准测试费用(美元)
    $23.0
  3. OpenAIGPT-5.6 Luna (low)
    智能得分 (Intelligence Index)
    21.00
    基准测试费用(美元)
    $27.0
  4. OpenAIGPT-5.6 Luna (medium)
    智能得分 (Intelligence Index)
    25.00
    基准测试费用(美元)
    $44.3
  5. OpenAIGPT-5.6 Luna (high)
    智能得分 (Intelligence Index)
    32.10
    基准测试费用(美元)
    $108
  6. OpenAIGPT-5.6 Luna (xhigh)
    智能得分 (Intelligence Index)
    34.60
    基准测试费用(美元)
    $179
  7. Xiaomi可下载权重的模型 (open weights)MiMo-V2.6-Pro
    智能得分 (Intelligence Index)
    46.30
    基准测试费用(美元)
    $207
  8. MetaMuse Spark 1.3 (max)
    智能得分 (Intelligence Index)
    48.10
    基准测试费用(美元)
    $2000
  9. OpenAIGPT-6 Astra (medium)
    智能得分 (Intelligence Index)
    49.60
    基准测试费用(美元)
    $2434
  10. OpenAIGPT-6 Astra (high)
    智能得分 (Intelligence Index)
    50.90
    基准测试费用(美元)
    $2925
  11. OpenAIGPT-6 Astra (xhigh)
    智能得分 (Intelligence Index)
    52.40
    基准测试费用(美元)
    $3803
  12. OpenAIGPT-6 Astra (max)
    智能得分 (Intelligence Index)
    52.70
    基准测试费用(美元)
    $5324
  13. AnthropicClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)
    智能得分 (Intelligence Index)
    53.20
    基准测试费用(美元)
    $9063
  14. AnthropicClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)
    智能得分 (Intelligence Index)
    53.40
    基准测试费用(美元)
    $13129
Last 30 天

最新发布

最新发布

650型号,性价比第一

#模型开发者智能得分 (Intelligence Index)基准测试费用(美元)每美元测试费用的智能得分端到端总耗时(E2E)可下载权重发布日期最佳性价比前沿
1GPT-5.6 Luna (low)OpenAI21.00$27.00.785.25s2026-07-09最佳性价比前沿
2GPT-5.6 Luna (Non-reasoning)OpenAI15.50$23.00.674.13s2026-07-09最佳性价比前沿
3Granite 4.2 3BIBM9.10$14.90.6111.7s2026-08-25最佳性价比前沿
4GPT-5.6 Luna (medium)OpenAI25.00$44.30.566.13s2026-07-09最佳性价比前沿
5DeepSeek V3 (Dec '24)DeepSeek8.50$25.40.34-2024-12-26
6gpt-oss-20b (high)OpenAI9.00$29.40.3115.5s2025-08-05
7GPT-5.6 Luna (high)OpenAI32.10$1080.3013.3s2026-07-09最佳性价比前沿
8MiMo-V2.6-ProXiaomi46.30$2070.2222.4s2026-09-21最佳性价比前沿
9Granite 4.2 8BIBM11.10$53.10.2129.7s2026-08-25开放权重前沿
10Mistral Small 3.1Mistral7.10$35.80.204.12s2025-03-17
11GPT-5.6 Luna (xhigh)OpenAI34.60$1790.1940.3s2026-07-09最佳性价比前沿
12MiMo-V2.5-ProXiaomi26.00$1380.1955.8s2026-04-22开放权重前沿
13Hy3Tencent25.30$1560.1630.4s2026-07-06
14GLM 5.3 FlashZ AI41.80$2800.1530.6s2026-08-26
15Mistral Small 4 (Reasoning)Mistral11.30$78.70.1416.8s2026-03-16开放权重前沿
16NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA8.90$63.70.1420.3s2025-12-15
17Gemini 3.1 Flash-LiteGoogle15.60$1190.137.38s2026-03-03
18DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek24.20$1910.13-2026-04-24
19Nemotron 3.5 LightningNVIDIA12.90$1050.129.02s2026-08-11开放权重前沿
20GPT-5.6 Luna (max)OpenAI37.30$3200.12126.5s2026-07-09
21Muse Glimmer (high)Meta17.50$1520.1227.3s2026-08-10
22DeepSeek V3 0324DeepSeek9.70$87.20.11-2025-03-25
23Qwen3.8-Flash-NextAlibaba39.80$3630.1146.5s2026-08-26
24DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek30.40$2770.1153.5s2026-04-24
25Celeris-1Celeris6.30$60.00.110.92s2026-07-24
26MiniMax-M2.7MiniMax22.80$2360.1038.6s2026-03-18
27gpt-oss-120b (high)OpenAI11.60$1220.0912.9s2025-08-05
28GPT-5 mini (high)OpenAI16.80$1890.09108.3s2025-08-07
29LongCat 2.0LongCat19.10$2210.09-2026-06-29
30Gemini 3.5 Flash-LiteGoogle22.20$2670.0812.5s2026-07-21
31DeepSeek V4.1 Flash (Reasoning, Max Effort)DeepSeek39.50$4770.0812.1s2026-09-10
32GPT-5.6 Terra (Non-reasoning)OpenAI20.80$2560.086.81s2026-07-09
33GPT-5.6 Terra (low)OpenAI27.50$3500.087.72s2026-07-09
34Ministral 3 3BMistral4.80$61.30.082.98s2025-12-02
35DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek34.80$4450.0811.8s2026-08-21
36Grok 4.3 (high)SpaceXAI24.90$3320.0732.3s2026-04-30
37DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek34.30$4740.0712.2s2026-07-31
38Mistral Large 3Mistral9.30$1300.077.71s2025-12-02
39Grok 4.3 (Non-reasoning)SpaceXAI14.00$2030.075.56s2026-04-30
40GPT-5.6 Terra (medium)OpenAI30.10$4470.077.61s2026-07-09
41GPT-5.4 nano (xhigh)OpenAI20.70$3310.06114.7s2026-03-17
42MiniMax-M3MiniMax29.20$5380.0517.5s2026-06-01
43Qwen3 235B A22B 2507 (Reasoning)Alibaba12.70$2350.0539.7s2025-07-25
44GPT-5.6 Sol (low)OpenAI33.50$6370.0510.7s2026-07-09
45Qwen3 30B A3B 2507 (Reasoning)Alibaba9.80$1880.0519.7s2025-07-30
46Solar Pro 3Upstage7.80$1590.0519.7s2026-04-06
47Step 5 PreviewStepFun43.70$9250.0533.7s2026-09-18
48Grok 4.6 (low)SpaceXAI35.10$7610.0516.9s2026-08-12
49GPT-5.6 Terra (high)OpenAI34.20$7680.048.87s2026-07-09
50Qwen3.7 PlusAlibaba25.20$5680.0438.7s2026-06-01
51GPT-5.6 Sol (medium)OpenAI39.20$9970.0412.5s2026-07-09
52Claude Sonnet 5 (Adaptive Reasoning, Low Effort)Anthropic24.30$6530.049.67s2026-06-30
53Gemini 3.8 Flash (medium)Google39.80$11000.04-2026-09-02
54Gemini 3.7 Flash (high)Google39.10$10840.0413.5s2026-08-13
55Gemini 3.6 Flash (high)Google34.00$10370.0321.5s2026-07-21
56Ministral 3 8BMistral5.50$1690.037.06s2025-12-02
57Claude 4.5 Haiku (Reasoning)Anthropic16.90$5240.0326.4s2025-10-15
58DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek36.00$11220.0335.6s2026-08-13
59GPT-5.6 Terra (xhigh)OpenAI38.00$11870.0323.7s2026-07-09
60Qwen3.8 27B (low)Alibaba26.20$8760.0351.2s2026-08-14
61GPT-6 Astra (low)OpenAI45.80$15370.0312.0s2026-09-03
62Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA22.90$7780.0318.3s2026-06-04
63Ring-2.6-1TInclusionAI16.60$5760.0322.9s2026-05-08
64Qwen3.8 27B (xhigh)Alibaba33.70$11700.0361.2s2026-08-14
65Muse Spark 1.2 (xhigh)Meta39.60$13850.0330.7s2026-08-05
66GPT-5.6 Sol (high)OpenAI42.30$14870.0316.5s2026-07-09
67Qwen3.8 27B (medium)Alibaba27.60$9770.0351.3s2026-08-14
68Grok 4.5 (high)SpaceXAI38.80$14040.0317.2s2026-07-08
69Muse Spark 1.3 (xhigh)Meta45.10$16550.0331.9s2026-09-02
70Kimi K2.7 CodeKimi25.80$9930.0347.3s2026-06-12
71Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic39.40$15610.0311.4s2026-07-24
72Gemini 3.8 Flash (high)Google40.90$16230.0316.9s2026-09-02
73DeepSeek R1 (Jan '25)DeepSeek11.40$4610.02-2025-01-20
74Inkling (xhigh)Thinking Machines25.00$10260.0227.4s2026-07-15
75Muse Spark 1.3 (max)Meta48.10$20000.0232.1s2026-09-02最佳性价比前沿
76Ministral 3 14BMistral6.00$2500.026.26s2025-12-02
77Qwen3.5 122B A10B (Reasoning)Alibaba15.60$6610.0221.8s2026-02-24
78Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)Anthropic28.10$12180.029.99s2026-06-30
79Gemini 2.5 ProGoogle16.10$7070.0227.8s2025-06-05
80Gemini 3.1 Pro PreviewGoogle29.70$13100.0242.0s2026-02-19
81Mistral Medium 3.1Mistral9.20$4060.025.93s2025-08-12
82Grok 4.6 (medium)SpaceXAI42.80$19370.0243.2s2026-08-12
83GPT-5.5 Instant (June 2026)OpenAI26.00$11840.0218.1s2026-06-25
84GLM-5.2 (max)Z AI33.70$15590.0240.6s2026-06-16
85GPT-5.6 Sol (xhigh)OpenAI44.00$20820.0239.9s2026-07-09
86Trinity Large ThinkingArcee AI10.80$5180.029.39s2026-04-01
87GPT-6 Astra (medium)OpenAI49.60$24340.0215.0s2026-09-03最佳性价比前沿
88Qwen3.6 35B A3B (Reasoning)Alibaba18.20$8960.0252.0s2026-04-16
89Qwen3.5 397B A17B (Reasoning)Alibaba18.40$9410.0243.7s2026-02-16
90Muse Spark 1.1 (xhigh)Meta33.70$17370.02-2026-07-09
91Qwen3 Coder NextAlibaba9.20$4760.026.95s2026-02-03
92Grok 4.6 (high)SpaceXAI44.30$23520.0251.0s2026-08-12
93Qwen3.6 27B (Reasoning)Alibaba21.40$11450.02108.6s2026-04-22
94GLM-5.1 (Reasoning)Z AI26.10$14180.02118.0s2026-04-07
95GLM-5.3 (max)Z AI44.80$25030.0244.1s2026-08-18
96Kimi K2.6Kimi27.00$15280.02144.4s2026-04-20
97GPT-6 Astra (high)OpenAI50.90$29250.0267.9s2026-09-03最佳性价比前沿
98GPT-5.6 Terra (max)OpenAI42.10$25010.02241.9s2026-07-09
99GPT-5.4 mini (xhigh)OpenAI24.10$14380.02107.3s2026-03-17
100Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic44.80$27320.0213.5s2026-07-24
101GPT-5.5 (medium)OpenAI33.80$20950.0212.3s2026-04-23
102Grok 4.6 (xhigh)SpaceXAI44.20$28300.0260.1s2026-08-12
103Claude Sonnet 5 (Adaptive Reasoning, High Effort)Anthropic31.70$20860.0218.8s2026-06-30
104Qwen3.8 2.4T A95BAlibaba39.90$26330.0265.5s2026-08-12
105Gemini 3.5 Flash (high)Google32.60$21720.0226.9s2026-05-19
106Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic46.80$31580.0116.9s2026-09-01
107Claude 4.5 Sonnet (Reasoning)Anthropic20.70$15010.0123.7s2025-09-29
108GPT-6 Astra (xhigh)OpenAI52.40$38030.01131.8s2026-09-03最佳性价比前沿
109GPT-5.6 Sol (max)OpenAI47.00$34650.01130.2s2026-07-09
110Qwen3.7 MaxAlibaba29.50$22770.0117.1s2026-05-19
111Qwen3.8 27B (Non-reasoning)Alibaba20.20$15700.0113.8s2026-08-14
112Qwen3.8 MaxAlibaba40.20$31720.0166.1s2026-08-03
113Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic48.90$39830.0117.4s2026-09-01
114Mistral Medium 3.5Mistral14.20$11600.0119.8s2026-04-29
115Grok 4.7 (high)SpaceXAI46.30$38810.0110.1s2026-09-21
116Kimi K3 (max)Kimi43.60$36580.0169.5s2026-07-16
117Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic48.10$43320.0123.7s2026-07-24
118GPT-5.5 (high)OpenAI37.00$34110.0123.0s2026-04-23
119Quasar 438B (max, based on GLM-5.2)Multiverse Computing26.70$25040.0117.5s2026-08-10
120Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)Anthropic34.40$32550.0164.4s2026-06-30
121GPT-6 Astra (max)OpenAI52.70$53240.01283.7s2026-09-03最佳性价比前沿
122Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic51.20$52420.0132.5s2026-09-01
123Nemotron 3 Super 120B A12B (Reasoning)NVIDIA12.80$13190.0116.4s2026-03-11
124Grok 4.7 (xhigh)SpaceXAI46.40$49670.0113.6s2026-09-21
125Qwen3.8 Max (0902)Alibaba45.40$49350.0166.4s2026-09-02
126Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic49.70$58680.0130.8s2026-07-24
127GPT-5.5 (xhigh)OpenAI38.40$52940.0175.8s2026-04-23
128Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic50.80$72750.0158.1s2026-07-24
129Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic30.10$49250.01149.9s2026-02-17
130Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic41.80$68740.0139.6s2026-05-28
131Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic53.20$90630.01211.2s2026-09-01最佳性价比前沿
132Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic38.20$69980.01149.1s2026-06-30
133Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic49.60$111610.00142.8s2026-06-09
134Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic53.40$131290.00298.9s2026-09-01最佳性价比前沿
135A.X-K2SK Telecom22.70---2026-08-12
136Agnes 2.5 Pro AlphaSapiens AI26.80--17.4s2026-07-24
137Agnes 2.5 Pro BetaSapiens AI35.20---2026-08-26
138Agnes 3.0 FlashSapiens AI35.50---2026-09-11
139Apertus 70B InstructSwiss AI Initiative5.10---2025-09-02
140Apertus 8B InstructSwiss AI Initiative4.80---2025-09-02
141Apodex 1.1Apodex30.40---2026-08-30
142Apriel-v1.5-15B-ThinkerServiceNow13.80---2025-09-30
143Apriel-v1.6-15B-ThinkerServiceNow13.40---2025-11-25
144Arctic InstructSnowflake5.40---2024-04-24
145Claude 2.0Anthropic5.50---2023-07-11
146Claude 2.1Anthropic5.60---2023-11-21
147Claude 3 HaikuAnthropic5.60---2024-03-04
148Claude 3 OpusAnthropic8.70---2024-03-04
149Claude 3 SonnetAnthropic5.90---2024-03-04
150Claude 3.5 HaikuAnthropic8.90---2024-10-22
151Claude 3.5 Sonnet (June '24)Anthropic7.20---2024-06-21
152Claude 3.5 Sonnet (Oct '24)Anthropic7.90---2024-10-22
153Claude 3.7 Sonnet (Non-reasoning)Anthropic15.30---2025-02-24
154Claude 3.7 Sonnet (Reasoning)Anthropic17.70---2025-02-24
155Claude 4 Opus (Non-reasoning)Anthropic16.60---2025-05-22
156Claude 4 Opus (Reasoning)Anthropic20.60---2025-05-22
157Claude 4 Sonnet (Non-reasoning)Anthropic16.60---2025-05-22
158Claude 4 Sonnet (Reasoning)Anthropic18.90---2025-05-22
159Claude 4.1 Opus (Non-reasoning)Anthropic18.60---2025-08-05
160Claude 4.1 Opus (Reasoning)Anthropic22.80---2025-08-05
161Claude 4.5 Haiku (Non-reasoning)Anthropic15.40--6.36s2025-10-15
162Claude 4.5 Sonnet (Non-reasoning)Anthropic19.30--13.6s2025-09-29
163Claude InstantAnthropic5.00---2023-03-14
164Claude Opus 4.5 (Non-reasoning)Anthropic23.70--11.8s2025-11-24
165Claude Opus 4.5 (Reasoning)Anthropic29.10--23.1s2025-11-24
166Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic31.90--35.9s2026-02-05
167Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic26.40--14.3s2026-02-05
168Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic40.70--33.8s2026-04-16
169Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic30.90--12.6s2026-04-16
170Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic24.70--12.3s2026-02-17
171Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic23.30--12.8s2026-02-17
172Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic23.20--8.85s2026-06-30
173Cogito v2.1 (Reasoning)Deep Cogito----2025-11-18
174Command ACohere7.00--10.0s2025-03-13
175Command-R (Mar '24)Cohere5.00---2024-03-12
176Command-R+ (Apr '24)Cohere5.30---2024-04-04
177DBRX InstructDatabricks5.30---2024-03-27
178DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research5.10---2025-02-13
179DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research6.10---2025-03-13
180DeepSeek Coder V2 Lite InstructDeepSeek5.30---2024-06-17
181DeepSeek LLM 67B Chat (V1)DeepSeek5.30---2023-11-29
182DeepSeek R1 0528 (May '25)DeepSeek13.10---2025-05-28
183DeepSeek R1 0528 Qwen3 8BDeepSeek8.10---2025-05-29
184DeepSeek R1 Distill Llama 70BDeepSeek7.90---2025-01-20
185DeepSeek R1 Distill Llama 8BDeepSeek6.50---2025-01-20
186DeepSeek R1 Distill Qwen 1.5BDeepSeek5.50---2025-01-20
187DeepSeek R1 Distill Qwen 14BDeepSeek7.80---2025-01-20
188DeepSeek R1 Distill Qwen 32BDeepSeek8.40---2025-01-20
189DeepSeek V3.1 (Non-reasoning)DeepSeek13.70---2025-08-21
190DeepSeek V3.1 (Reasoning)DeepSeek13.50---2025-08-21
191DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek13.90---2025-09-22
192DeepSeek V3.1 Terminus (Reasoning)DeepSeek14.80---2025-09-22
193DeepSeek V3.2 (Non-reasoning)DeepSeek16.00---2025-12-01
194DeepSeek V3.2 (Reasoning)DeepSeek21.50---2025-12-01
195DeepSeek V3.2 Exp (Non-reasoning)DeepSeek13.90---2025-09-29
196DeepSeek V3.2 Exp (Reasoning)DeepSeek16.60---2025-09-29
197DeepSeek V3.2 SpecialeDeepSeek14.50---2025-12-01
198DeepSeek V4 Flash 0420 (Non-reasoning)DeepSeek18.90---2026-04-24
199DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek26.00---2026-04-24
200DeepSeek V4 Pro 0424 (Non-reasoning)DeepSeek20.80--6.75s2026-04-24
201DeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeek30.10--31.3s2026-04-24
202DeepSeek-Coder-V2DeepSeek6.00---2024-06-17
203DeepSeek-V2-ChatDeepSeek5.50---2024-05-06
204DeepSeek-V2.5DeepSeek6.60---2024-09-06
205DeepSeek-V2.5 (Dec '24)DeepSeek6.60---2024-12-10
206Devstral MediumMistral9.00---2025-07-10
207Devstral Small (Jul '25)Mistral7.60---2025-07-10
208Devstral Small (May '25)Mistral8.70---2025-05-21
209DiffusionGemma 26B A4BGoogle9.50---2026-06-10
210Doubao Seed CodeByteDance Seed16.90---2025-11-11
211ERNIE 4.5 300B A47BBaidu7.50---2025-06-30
212ERNIE 5.0 Thinking PreviewBaidu14.30---2025-11-13
213Exaone 4.0 1.2B (Non-reasoning)LG AI Research5.20---2025-07-15
214Exaone 4.0 1.2B (Reasoning)LG AI Research5.30---2025-07-15
215EXAONE 4.0 32B (Non-reasoning)LG AI Research6.30---2025-07-15
216EXAONE 4.0 32B (Reasoning)LG AI Research8.20---2025-07-15
217EXAONE 4.5 33BLG AI Research13.20---2026-04-09
218EXAONE 4.5 33B (Non-reasoning)LG AI Research----2026-04-09
219Falcon-H1R-7BTII UAE7.80---2026-01-04
220G9v3-39A5BAI9Stars21.80---2026-08-20
221G9v3-3BAI9Stars10.80---2026-07-23
222Gemini 1.0 ProGoogle5.30---2023-12-06
223Gemini 1.0 UltraGoogle5.80---2023-12-06
224Gemini 1.5 Flash (May '24)Google5.90---2024-05-14
225Gemini 1.5 Flash (Sep '24)Google7.10---2024-09-24
226Gemini 1.5 Flash-8BGoogle6.20---2024-10-03
227Gemini 1.5 Pro (May '24)Google6.40---2024-05-15
228Gemini 1.5 Pro (Sep '24)Google7.90---2024-09-24
229Gemini 2.0 Flash (experimental)Google8.20---2024-12-11
230Gemini 2.0 Flash (Feb '25)Google8.90---2025-02-05
231Gemini 2.0 Flash Thinking Experimental (Dec '24)Google6.60---2024-12-19
232Gemini 2.0 Flash Thinking Experimental (Jan '25)Google9.40---2025-01-21
233Gemini 2.0 Flash-Lite (Feb '25)Google7.40---2025-02-25
234Gemini 2.0 Flash-Lite (Preview)Google7.30---2025-02-05
235Gemini 2.0 Pro Experimental (Feb '25)Google8.70---2025-02-05
236Gemini 2.5 Flash (Non-reasoning)Google9.90--2.80s2025-05-20
237Gemini 2.5 Flash (Reasoning)Google13.10--19.3s2025-05-20
238Gemini 2.5 Flash Preview (Non-reasoning)Google8.70---2025-04-17
239Gemini 2.5 Flash Preview (Reasoning)Google11.70---2025-04-17
240Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google12.40---2025-09-25
241Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google15.50---2025-09-25
242Gemini 2.5 Flash-Lite (Non-reasoning)Google6.70--1.88s2025-06-17
243Gemini 2.5 Flash-Lite (Reasoning)Google8.50--22.2s2025-06-17
244Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google9.30---2025-09-25
245Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google10.40---2025-09-25
246Gemini 2.5 Pro Preview (Mar' 25)Google15.00---2025-03-25
247Gemini 2.5 Pro Preview (May' 25)Google14.50---2025-05-06
248Gemini 3 Deep ThinkGoogle----2026-02-05
249Gemini 3 Flash Preview (Non-reasoning)Google17.90--3.29s2025-12-17
250Gemini 3 Flash Preview (Reasoning)Google26.30--8.33s2025-12-17
251Gemini 3 Pro Preview (high)Google28.00---2025-11-18
252Gemini 3 Pro Preview (low)Google22.30---2025-11-18
253Gemini 3.5 Flash (medium)Google33.60--17.9s2026-05-19
254Gemini 3.5 Flash (minimal)Google23.80--3.52s2026-05-19
255Gemini 3.7 Flash (low)Google36.90--2.35s2026-08-13
256Gemini 3.7 Flash (medium)Google39.60--6.14s2026-08-13
257Gemini 3.8 Flash (low)Google33.50---2026-09-02
258Gemma 3 1B InstructGoogle4.80---2025-03-13
259Gemma 3 270MGoogle5.10---2025-08-14
260Gemma 3 4B InstructGoogle4.80---2025-03-12
261Gemma 3n E2B InstructGoogle4.80---2025-06-26
262Gemma 3n E4B InstructGoogle4.80---2025-06-26
263Gemma 3n E4B Instruct Preview (May '25)Google5.80---2025-05-20
264Gemma 4 12B (Non-reasoning)Google9.40--6.71s2026-06-03
265Gemma 4 12B (Reasoning)Google14.20--24.5s2026-06-03
266Gemma 4 26B A4B (Non-reasoning)Google13.10--8.02s2026-04-02
267Gemma 4 26B A4B (Reasoning)Google16.70---2026-04-02
268Gemma 4 31B (Non-reasoning)Google13.90--17.1s2026-04-02
269Gemma 4 31B (Reasoning)Google19.00--64.2s2026-04-02
270Gemma 4 E2B (Non-reasoning)Google6.50---2026-04-02
271Gemma 4 E2B (Reasoning)Google7.80---2026-04-02
272Gemma 4 E4B (Non-reasoning)Google7.50--11.9s2026-04-03
273Gemma 4 E4B (Reasoning)Google8.90--55.4s2026-04-03
274GLM 5V Turbo (Reasoning)Z AI23.50---2026-04-01
275GLM-4.5 (Reasoning)Z AI12.80---2025-07-28
276GLM-4.5-AirZ AI11.10--33.5s2025-07-28
277GLM-4.5V (Non-reasoning)Z AI6.70--18.3s2025-08-11
278GLM-4.5V (Reasoning)Z AI7.60--74.9s2025-08-11
279GLM-4.6 (Non-reasoning)Z AI14.90--18.5s2025-09-30
280GLM-4.6 (Reasoning)Z AI18.50--73.1s2025-09-30
281GLM-4.6V (Non-reasoning)Z AI8.40--9.82s2025-12-08
282GLM-4.6V (Reasoning)Z AI11.20--48.8s2025-12-08
283GLM-4.7 (Non-reasoning)Z AI17.40--6.12s2025-12-22
284GLM-4.7 (Reasoning)Z AI22.20--25.1s2025-12-22
285GLM-4.7-Flash (Non-reasoning)Z AI10.60--5.72s2026-01-19
286GLM-4.7-Flash (Reasoning)Z AI14.90--30.6s2026-01-19
287GLM-5 (Non-reasoning)Z AI21.80--8.78s2026-02-11
288GLM-5 (Reasoning)Z AI27.90--50.1s2026-02-11
289GLM-5-TurboZ AI26.60---2026-03-15
290GLM-5.1 (Non-reasoning)Z AI24.20--14.1s2026-04-07
291GLM-5.2 (Non-reasoning)Z AI22.40--4.11s2026-06-16
292GPT-3.5 TurboOpenAI5.50---2022-11-30
293GPT-3.5 Turbo (0613)OpenAI----2023-06-13
294GPT-4OpenAI6.70---2023-03-14
295GPT-4 TurboOpenAI7.00--19.0s2023-11-06
296GPT-4.1OpenAI12.70--3.62s2025-04-14
297GPT-4.1 miniOpenAI10.20--4.31s2025-04-14
298GPT-4.1 nanoOpenAI7.80--3.56s2025-04-14
299GPT-4.5 (Preview)OpenAI9.60---2025-02-27
300GPT-4o (Aug '24)OpenAI7.70--5.48s2024-08-06
301GPT-4o (ChatGPT)OpenAI7.20---2025-02-15
302GPT-4o (March 2025, chatgpt-4o-latest)OpenAI9.00---2025-03-27
303GPT-4o (May '24)OpenAI7.30--5.27s2024-05-13
304GPT-4o (Nov '24)OpenAI8.40--4.26s2024-11-20
305GPT-4o miniOpenAI6.70--4.78s2024-07-18
306GPT-4o mini Realtime (Dec '24)OpenAI----2024-12-17
307GPT-4o Realtime (Dec '24)OpenAI----2024-12-17
308GPT-5 (ChatGPT)OpenAI10.40---2025-08-07
309GPT-5 (high)OpenAI23.00--72.5s2025-08-07
310GPT-5 (low)OpenAI20.80--12.9s2025-08-07
311GPT-5 (medium)OpenAI22.90--33.0s2025-08-07
312GPT-5 (minimal)OpenAI11.40--6.82s2025-08-07
313GPT-5 Codex (high)OpenAI24.90---2025-09-23
314GPT-5 mini (medium)OpenAI20.60--18.7s2025-08-07
315GPT-5 mini (minimal)OpenAI9.90--5.97s2025-08-07
316GPT-5 nano (high)OpenAI13.00--103.1s2025-08-07
317GPT-5 nano (medium)OpenAI12.50--48.2s2025-08-07
318GPT-5 nano (minimal)OpenAI7.10--3.67s2025-08-07
319GPT-5.1 (high)OpenAI24.70--32.8s2025-11-13
320GPT-5.1 (Non-reasoning)OpenAI13.30--4.88s2025-11-13
321GPT-5.1 Codex (high)OpenAI23.70---2025-11-13
322GPT-5.1 Codex mini (high)OpenAI20.40---2025-11-13
323GPT-5.2 (medium)OpenAI26.50---2025-12-11
324GPT-5.2 (Non-reasoning)OpenAI17.00--7.19s2025-12-11
325GPT-5.2 (xhigh)OpenAI30.40--200.5s2025-12-11
326GPT-5.2 Codex (xhigh)OpenAI28.50---2025-12-11
327GPT-5.3 Codex (xhigh)OpenAI32.50--64.3s2026-02-05
328GPT-5.4 (low)OpenAI27.60--6.15s2026-03-05
329GPT-5.4 (Non-reasoning)OpenAI18.20--5.82s2026-03-05
330GPT-5.4 (xhigh)OpenAI39.00--154.5s2026-03-05
331GPT-5.4 mini (medium)OpenAI19.70--16.5s2026-03-17
332GPT-5.4 mini (Non-Reasoning)OpenAI11.10--3.40s2026-03-17
333GPT-5.4 nano (medium)OpenAI20.00--7.04s2026-03-17
334GPT-5.4 nano (Non-Reasoning)OpenAI11.70--3.77s2026-03-17
335GPT-5.4 Pro (xhigh)OpenAI----2026-03-05
336GPT-5.5 (low)OpenAI30.70--6.83s2026-04-23
337GPT-5.5 (Non-reasoning)OpenAI23.20--7.24s2026-04-23
338GPT-5.5 Instant (May 2026)OpenAI22.70---2026-05-05
339GPT-5.5 Pro (xhigh)OpenAI----2026-04-23
340GPT-5.6 Sol (Non-reasoning)OpenAI28.30--8.90s2026-07-09
341gpt-oss-120b (low)OpenAI10.20--12.8s2025-08-05
342gpt-oss-20b (low)OpenAI10.00--12.4s2025-08-05
343Granite 3.3 8B (Non-reasoning)IBM4.90--56.1s2025-04-16
344Granite 4.0 1BIBM5.00---2025-10-28
345Granite 4.0 350MIBM4.80---2025-10-28
346Granite 4.0 H 1BIBM5.20---2025-10-28
347Granite 4.0 H 350MIBM4.80---2025-10-28
348Granite 4.0 H SmallIBM6.00--41.9s2025-09-22
349Granite 4.0 MicroIBM5.10---2025-09-22
350Granite 4.1 30BIBM7.40---2026-04-29
351Granite 4.1 3BIBM5.90---2026-04-29
352Granite 4.1 8BIBM6.60--6.88s2026-04-29
353Granite 4.2 30BIBM14.80--35.2s2026-08-25
354Grok 2 (Dec '24)SpaceXAI7.10---2024-12-12
355Grok 3SpaceXAI12.10---2025-02-19
356Grok 3 mini Reasoning (high)SpaceXAI14.60--33.4s2025-02-19
357Grok 3 Reasoning BetaSpaceXAI10.40---2025-02-19
358Grok 4SpaceXAI22.50---2025-07-10
359Grok 4 Fast (Non-reasoning)SpaceXAI11.10---2025-09-19
360Grok 4 Fast (Reasoning)SpaceXAI17.90---2025-09-19
361Grok 4.1 Fast (Non-reasoning)SpaceXAI11.30---2025-11-19
362Grok 4.1 Fast (Reasoning)SpaceXAI20.40---2025-11-19
363Grok 4.20 0309 (Non-reasoning)SpaceXAI14.60---2026-03-10
364Grok 4.20 0309 (Reasoning)SpaceXAI25.20---2026-03-10
365Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI14.20--6.30s2026-04-07
366Grok 4.20 0309 v2 (Reasoning)SpaceXAI25.70--31.9s2026-04-07
367Grok 4.3 (low)SpaceXAI24.30--12.2s2026-04-30
368Grok 4.3 (medium)SpaceXAI24.80--17.7s2026-04-30
369Grok BetaSpaceXAI6.90---2024-08-13
370Grok Build 0.1 0616SpaceXAI27.20--50.3s2026-06-16
371Grok Code Fast 1SpaceXAI14.10---2025-08-28
372Grok-1SpaceXAI6.30---2024-03-17
373Hermes 3 - Llama-3.1 70BNous Research6.00--18.8s2024-08-15
374Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research7.40--14.5s2025-08-27
375Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research7.50--64.7s2025-08-27
376Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research6.70---2025-08-27
377Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research7.90---2025-08-27
378Hy3-preview (Non-reasoning)Tencent17.00---2026-04-23
379Hy3-preview (Reasoning)Tencent22.70---2026-04-23
380HyperCLOVA X SEED Think (32B)Naver11.40---2025-12-26
381HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing11.70---2026-05-26
382Inkling SmallThinking Machines27.80--14.9s2026-07-30
383INTELLECT-3Prime Intellect10.60---2025-11-27
384Jamba 1.5 LargeAI21 Labs6.00---2024-08-22
385Jamba 1.5 MiniAI21 Labs5.20---2024-08-22
386Jamba 1.6 LargeAI21 Labs6.00---2025-03-06
387Jamba 1.6 MiniAI21 Labs5.20---2025-03-06
388Jamba 1.7 LargeAI21 Labs6.10---2025-07-07
389Jamba 1.7 MiniAI21 Labs5.20---2025-07-07
390Jamba Reasoning 3BAI21 Labs5.70---2025-10-08
391JT-35B-FlashChina Mobile18.70---2026-05-14
392JT-4.1 Flash 236B A21BChina Mobile27.30---2026-07-09
393JT-MINIChina Mobile12.20---2026-04-15
394K-EXAONE (Non-reasoning)LG AI Research11.20---2025-12-31
395K-EXAONE (Reasoning)LG AI Research14.40---2025-12-31
396K-EXAONE 2.0 0803LG AI Research19.70---2026-08-12
397K2 Horizon 0.9BFoundation Models的Institute3.00---2026-09-03
398K2 Horizon 3.7BFoundation Models的Institute15.60---2026-09-03
399K2 Horizon 375B A23BFoundation Models的Institute30.50---2026-09-03
400K2 Horizon 7BFoundation Models的Institute20.60---2026-09-03
401K2 Horizon MoVA 36B A4BFoundation Models的Institute25.30---2026-09-03
402K2 Think V2Foundation Models的Institute11.50---2025-12-15
403K2-V2 (high)Foundation Models的Institute9.90---2025-12-05
404K2-V2 (low)Foundation Models的Institute7.30---2025-12-05
405K2-V2 (medium)Foundation Models的Institute9.00---2025-12-05
406KAT Coder Pro V2KwaiKAT21.70---2026-03-27
407KAT-Coder-Pro V1KwaiKAT18.60---2025-11-11
408Kimi K2Kimi12.70--12.2s2025-07-11
409Kimi K2 0905Kimi15.30--11.7s2025-09-05
410Kimi K2 ThinkingKimi22.00--21.9s2025-11-06
411Kimi K2.5 (Non-reasoning)Kimi19.40---2026-01-27
412Kimi K2.5 (Reasoning)Kimi23.50---2026-01-27
413Kimi K2.6 (Non-reasoning)Kimi23.60--16.4s2026-04-20
414Kimi K3 (low)Kimi34.50--71.6s2026-07-16
415Kimi Linear 48B A3B InstructKimi7.30---2025-10-30
416LFM 40BLiquid AI5.40---2024-09-30
417LFM2 1.2BLiquid AI4.80---2025-07-10
418LFM2 2.6BLiquid AI5.20---2025-09-23
419LFM2 24B A2BLiquid AI5.90---2026-02-25
420LFM2 8B A1BLiquid AI4.90---2025-10-07
421LFM2.5-1.2B-InstructLiquid AI5.20---2026-01-05
422LFM2.5-1.2B-ThinkingLiquid AI5.20---2026-01-20
423LFM2.5-2.6BLiquid AI8.40---2026-08-04
424LFM2.5-8B-A1BLiquid AI7.20---2026-05-28
425LFM2.5-VL-1.6BLiquid AI4.80---2026-01-05
426Ling 2.6 FlashInclusionAI9.70---2026-04-21
427Ling 3.0 FlashInclusionAI24.90--10.2s2026-08-04
428Ling 3.0 TinyInclusionAI15.30--69.5s2026-08-06
429Ling-1TInclusionAI9.20---2025-10-08
430Ling-2.6-1TInclusionAI17.00---2026-04-23
431Ling-3.0-flash-FinInclusionAI22.60--18.8s2026-09-11
432Ling-3.0-flash-VLInclusionAI24.60--20.0s2026-09-10
433Ling-flash-2.0InclusionAI7.80--97.7s2025-09-17
434Ling-mini-2.0InclusionAI5.50---2025-09-09
435Llama 2 Chat 13BMeta5.30---2023-07-18
436Llama 2 Chat 70BMeta5.30---2023-07-18
437Llama 2 Chat 7BMeta5.70---2023-07-18
438Llama 3 Instruct 70BMeta5.50---2024-04-18
439Llama 3 Instruct 8BMeta4.80---2024-04-18
440Llama 3.1 Instruct 405BMeta7.30---2024-07-23
441Llama 3.1 Instruct 70BMeta6.60--10.1s2024-07-23
442Llama 3.1 Instruct 8BMeta6.90--4.52s2024-07-23
443Llama 3.1 Nemotron Instruct 70BNVIDIA6.90--7.89s2024-10-15
444Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA7.30---2025-05-20
445Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA7.50---2025-04-07
446Llama 3.1 Tulu3 405BAllen Institute for AI7.20---2025-01-30
447Llama 3.2 Instruct 11B (Vision)Meta5.40--31.4s2024-09-25
448Llama 3.2 Instruct 1BMeta4.80---2024-09-25
449Llama 3.2 Instruct 3BMeta5.70---2024-09-25
450Llama 3.2 Instruct 90B (Vision)Meta6.40---2024-09-25
451Llama 3.3 Instruct 70BMeta7.70--7.77s2024-12-06
452Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA7.30---2025-03-18
453Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA8.90---2025-03-18
454Llama 4 MaverickMeta10.00--5.37s2025-04-05
455Llama 4 ScoutMeta8.10--5.89s2025-04-05
456Llama 65BMeta5.00---2023-02-24
457Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA7.40--8.86s2025-07-25
458Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA9.00--20.5s2025-07-25
459LongCat Flash LiteLongCat11.50---2026-01-28
460Magistral Medium 1Mistral9.10---2025-06-10
461Magistral Medium 1.2Mistral11.80---2025-09-18
462Magistral Small 1Mistral8.20---2025-06-10
463Magistral Small 1.2Mistral8.60---2025-09-17
464Mercury 2Inception13.80--5.45s2026-02-20
465Mi:dm K 2.5 ProKorea Telecom11.00---2025-12-11
466Mi:dm K 2.5 Pro PreviewKorea Telecom----2025-12-11
467MiMo-V2-Flash (Feb 2026)Xiaomi22.40---2025-12-16
468MiMo-V2-Flash (Non-reasoning)Xiaomi16.00---2025-12-16
469MiMo-V2-Flash (Reasoning)Xiaomi20.80---2025-12-16
470MiMo-V2-OmniXiaomi23.90---2026-03-19
471MiMo-V2-Omni-0327Xiaomi25.10---2026-03-27
472MiMo-V2-ProXiaomi28.60---2026-03-18
473MiMo-V2.5Xiaomi25.20--76.7s2026-04-22
474MiMo-V2.5-Pro (Non-reasoning)Xiaomi18.30--13.1s2026-04-22
475MiniCPM-V 4.6 1.3BOpenBMB5.70---2026-05-11
476MiniCPM5-1B (Non-reasoning)OpenBMB8.70---2026-05-25
477MiniCPM5-1B (Reasoning)OpenBMB8.80---2026-05-25
478MiniCPM5-2BOpenBMB12.50---2026-09-07
479MiniMax M1 40kMiniMax10.00---2025-06-17
480MiniMax M1 80kMiniMax11.70---2025-06-17
481MiniMax-M2MiniMax18.60--25.4s2025-10-26
482MiniMax-M2.1MiniMax20.90--26.3s2025-12-23
483MiniMax-M2.5MiniMax22.80--27.9s2026-02-12
484Mistral 7B InstructMistral5.00--6.43s2023-09-27
485Mistral Large (Feb '24)Mistral5.80---2024-02-26
486Mistral Large 2 (Jul '24)Mistral6.80---2024-07-24
487Mistral Large 2 (Nov '24)Mistral7.60---2024-11-18
488Mistral MediumMistral5.50--6.03s2023-12-11
489Mistral Medium 3Mistral9.00--5.72s2025-05-07
490Mistral SabaMistral6.50---2025-02-17
491Mistral Small (Feb '24)Mistral5.50--3.99s2024-02-26
492Mistral Small (Sep '24)Mistral5.80--4.28s2024-09-17
493Mistral Small 3Mistral6.70--4.31s2025-01-30
494Mistral Small 3.2Mistral8.20--3.93s2025-06-20
495Mistral Small 4 (Non-reasoning)Mistral9.00--4.00s2026-03-16
496Mixtral 8x22B InstructMistral5.70---2024-04-17
497Mixtral 8x7B InstructMistral5.10---2023-12-11
498Molmo 7B-DAllen Institute for AI5.60---2024-09-25
499Molmo2-8BAllen Institute for AI5.00---2025-12-11
500Motif 3Motif Technologies33.60---2026-08-12
501Motif 3 (Beta)Motif Technologies32.30---2026-07-14
502Motif-2-12.7B-ReasoningMotif Technologies9.20---2025-12-04
503Muse SparkMeta31.30---2026-04-08
504Nanbeige4.1-3BNanbeige8.40---2026-02-11
505Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA10.30--9.79s2026-04-29
506Nemotron Cascade 2 30B A3BNVIDIA11.70---2026-03-19
507Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI28.20---2026-06-02
508Nova 2.0 Lite (high)Amazon13.40--29.3s2025-10-29
509Nova 2.0 Lite (low)Amazon11.80--24.6s2025-10-29
510Nova 2.0 Lite (medium)Amazon12.50--31.3s2025-10-29
511Nova 2.0 Lite (Non-reasoning)Amazon8.70--4.24s2025-10-29
512Nova 2.0 Omni (low)Amazon11.10---2025-11-26
513Nova 2.0 Omni (medium)Amazon13.60---2025-11-26
514Nova 2.0 Omni (Non-reasoning)Amazon8.20---2025-11-26
515Nova 2.0 Pro Preview (low)Amazon12.80--33.9s2025-11-27
516Nova 2.0 Pro Preview (medium)Amazon14.20--36.1s2025-11-27
517Nova 2.0 Pro Preview (Non-reasoning)Amazon10.00--5.76s2025-11-27
518Nova LiteAmazon6.70--3.77s2024-12-03
519Nova MicroAmazon5.90--2.83s2024-12-03
520Nova PremierAmazon9.20---2025-04-30
521Nova ProAmazon7.00---2024-12-03
522NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA6.80--3.72s2025-12-15
523NVIDIA Nemotron 3 Nano 4BNVIDIA7.40---2026-03-16
524NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA5.80--4.19s2025-10-28
525NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA7.50--28.4s2025-10-28
526NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA6.80--4.75s2025-08-18
527NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA7.40--31.0s2025-08-18
528o1OpenAI15.20---2024-12-05
529o1-miniOpenAI9.80---2024-09-12
530o1-previewOpenAI11.40---2024-09-12
531o1-proOpenAI12.40---2025-03-19
532o3OpenAI20.20--10.5s2025-04-16
533o3-miniOpenAI12.50--8.88s2025-01-31
534o3-mini (high)OpenAI11.00--29.4s2025-01-31
535o3-proOpenAI21.90---2025-06-10
536o4-mini (high)OpenAI16.70--39.4s2025-04-16
537OLMo 2 32BAllen Institute for AI6.00---2025-03-13
538OLMo 2 7BAllen Institute for AI5.60---2024-11-26
539Olmo 3 32B ThinkAllen Institute for AI6.50---2025-11-20
540Olmo 3 7B InstructAllen Institute for AI5.20---2025-11-20
541Olmo 3 7B ThinkAllen Institute for AI5.60---2025-11-20
542Olmo 3.1 32B InstructAllen Institute for AI6.50---2026-01-13
543Olmo 3.1 32B ThinkAllen Institute for AI7.10---2025-12-12
544OpenChat 3.5 (1210)OpenChat5.30---2023-12-18
545PALM-2Google5.40---2023-05-10
546Phi-3 Mini Instruct 3.8BMicrosoft5.80---2024-04-23
547Phi-4Microsoft5.90--18.9s2024-12-12
548Phi-4 Mini InstructMicrosoft6.30--11.8s2024-02-26
549Phi-4 Multimodal InstructMicrosoft5.80--30.1s2025-02-26
550Pixtral LargeMistral7.10---2024-11-18
551Qwen Chat 14BAlibaba5.00---2023-09-25
552Qwen Chat 72BAlibaba5.40---2023-11-30
553Qwen1.5 Chat 110BAlibaba5.70---2024-04-25
554Qwen2 Instruct 72BAlibaba6.30---2024-06-07
555Qwen2.5 Coder Instruct 32BAlibaba6.70---2024-11-11
556Qwen2.5 Coder Instruct 7BAlibaba5.80---2024-09-19
557Qwen2.5 Instruct 32BAlibaba6.90---2024-09-19
558Qwen2.5 Instruct 72BAlibaba7.70---2024-09-19
559Qwen2.5 MaxAlibaba8.00---2025-01-28
560Qwen2.5 TurboAlibaba6.40--6.84s2024-11-18
561Qwen3 0.6B (Non-reasoning)Alibaba4.80---2025-04-28
562Qwen3 0.6B (Reasoning)Alibaba4.80---2025-04-28
563Qwen3 1.7B (Non-reasoning)Alibaba4.90---2025-04-28
564Qwen3 1.7B (Reasoning)Alibaba5.20---2025-04-28
565Qwen3 14B (Non-reasoning)Alibaba6.70--11.4s2025-04-28
566Qwen3 14B (Reasoning)Alibaba8.20--45.8s2025-04-28
567Qwen3 235B A22B (Non-reasoning)Alibaba8.30--12.1s2025-04-28
568Qwen3 235B A22B (Reasoning)Alibaba9.50--45.7s2025-04-28
569Qwen3 235B A22B 2507 InstructAlibaba12.00--11.3s2025-07-21
570Qwen3 30B A3B (Non-reasoning)Alibaba6.60--6.92s2025-04-28
571Qwen3 30B A3B (Reasoning)Alibaba7.60--27.5s2025-04-28
572Qwen3 30B A3B 2507 InstructAlibaba7.50--5.47s2025-07-29
573Qwen3 32B (Non-reasoning)Alibaba7.30--7.45s2025-04-28
574Qwen3 32B (Reasoning)Alibaba8.60--27.9s2025-04-28
575Qwen3 4B (Non-reasoning)Alibaba6.60---2025-04-28
576Qwen3 4B (Reasoning)Alibaba7.20---2025-04-28
577Qwen3 4B 2507 (Reasoning)Alibaba8.80---2025-08-06
578Qwen3 4B 2507 InstructAlibaba6.70---2025-08-06
579Qwen3 8B (Non-reasoning)Alibaba6.00--16.2s2025-04-28
580Qwen3 8B (Reasoning)Alibaba7.30--66.4s2025-04-28
581Qwen3 Coder 30B A3B InstructAlibaba9.60--8.64s2025-07-31
582Qwen3 Coder 480B A35B InstructAlibaba11.90--13.5s2025-07-22
583Qwen3 MaxAlibaba15.60--11.3s2025-09-23
584Qwen3 Max (Preview)Alibaba12.60--12.6s2025-09-05
585Qwen3 Max ThinkingAlibaba21.30---2026-01-26
586Qwen3 Max Thinking (Preview)Alibaba16.30--44.5s2025-11-03
587Qwen3 Next 80B A3B (Reasoning)Alibaba11.20--16.6s2025-09-11
588Qwen3 Next 80B A3B InstructAlibaba9.60--5.03s2025-09-11
589Qwen3 Omni 30B A3B (Reasoning)Alibaba7.80--27.1s2025-09-22
590Qwen3 Omni 30B A3B InstructAlibaba6.00--7.19s2025-09-22
591Qwen3 VL 235B A22B (Reasoning)Alibaba13.40--48.6s2025-09-23
592Qwen3 VL 235B A22B InstructAlibaba9.90--12.5s2025-09-23
593Qwen3 VL 30B A3B (Reasoning)Alibaba9.50--25.0s2025-10-03
594Qwen3 VL 30B A3B InstructAlibaba7.90--6.66s2025-10-03
595Qwen3 VL 32B (Reasoning)Alibaba11.90--31.0s2025-10-21
596Qwen3 VL 32B InstructAlibaba8.40--10.4s2025-10-21
597Qwen3 VL 4B (Reasoning)Alibaba7.00---2025-10-14
598Qwen3 VL 4B InstructAlibaba5.70---2025-10-14
599Qwen3 VL 8B (Reasoning)Alibaba8.20--25.2s2025-10-14
600Qwen3 VL 8B InstructAlibaba7.30--6.99s2025-10-14
601Qwen3.5 0.8B (Non-reasoning)Alibaba5.40---2026-03-02
602Qwen3.5 0.8B (Reasoning)Alibaba6.10---2026-03-02
603Qwen3.5 122B A10B (Non-reasoning)Alibaba17.70--5.79s2026-02-24
604Qwen3.5 27B (Non-reasoning)Alibaba19.40--11.8s2026-02-24
605Qwen3.5 27B (Reasoning)Alibaba22.90--38.8s2026-02-24
606Qwen3.5 2B (Non-reasoning)Alibaba6.20---2026-03-02
607Qwen3.5 2B (Reasoning)Alibaba6.90---2026-03-02
608Qwen3.5 35B A3B (Non-reasoning)Alibaba15.10--5.56s2026-02-24
609Qwen3.5 35B A3B (Reasoning)Alibaba19.30--18.6s2026-02-24
610Qwen3.5 397B A17B (Non-reasoning)Alibaba21.40--7.80s2026-02-16
611Qwen3.5 4B (Non-reasoning)Alibaba10.80--34.9s2026-03-02
612Qwen3.5 4B (Reasoning)Alibaba13.10--111.2s2026-03-02
613Qwen3.5 9B (Non-reasoning)Alibaba13.30--6.21s2026-03-02
614Qwen3.5 9B (Reasoning)Alibaba13.70--45.5s2026-03-02
615Qwen3.5 Omni FlashAlibaba12.50--4.00s2026-03-30
616Qwen3.5 Omni PlusAlibaba20.40--7.33s2026-03-30
617Qwen3.6 27B (Non-reasoning)Alibaba19.80--12.0s2026-04-22
618Qwen3.6 35B A3B (Non-reasoning)Alibaba15.20--6.14s2026-04-16
619Qwen3.6 Max PreviewAlibaba28.40--39.2s2026-04-20
620Qwen3.6 PlusAlibaba27.00--109.5s2026-04-02
621QwQ 32BAlibaba9.50---2025-03-05
622QwQ 32B-PreviewAlibaba7.60---2024-11-27
623R1 1776Perplexity6.40---2025-02-18
624Reka Flash (Sep '24)Reka AI6.40---2024-10-04
625Reka Flash 3Reka AI5.60---2025-03-10
626Ring-1TInclusionAI10.90---2025-10-13
627Ring-flash-2.0InclusionAI7.20---2025-09-19
628Sarvam 105B (high)Sarvam8.80---2026-03-06
629Sarvam 30B (high)Sarvam6.60---2026-03-06
630Sarvam M (Reasoning)Sarvam5.30---2025-05-23
631Seed-OSS-36B-InstructByteDance Seed12.10--65.1s2025-08-20
632Solar MiniUpstage6.40---2024-01-25
633Solar Open 100B (Reasoning)Upstage10.40---2025-12-17
634Solar Open2 250BUpstage24.70---2026-08-12
635Solar Pro 2 (Non-reasoning)Upstage7.00---2025-07-09
636Solar Pro 2 (Preview) (Non-reasoning)Upstage7.90---2025-05-20
637Solar Pro 2 (Preview) (Reasoning)Upstage9.10---2025-05-20
638Solar Pro 2 (Reasoning)Upstage7.50---2025-07-09
639Solar Pro 4Upstage28.20--39.8s2026-08-06
640SonarPerplexity7.70---2025-01-21
641Sonar ProPerplexity7.60---2025-01-21
642Sonar ReasoningPerplexity8.70---2025-01-28
643Sonar Reasoning ProPerplexity11.80---2025-01-28
644Step 3.5 FlashStepFun16.60--23.4s2026-02-02
645Step 3.5 Flash 2603StepFun17.00--25.2s2026-04-02
646Step 3.7 FlashStepFun19.50--15.2s2026-05-29
647Step3 VL 10BStepFun7.70---2026-01-20
648Tiny Aya GlobalCohere4.80---2026-02-17
649Tri-21B-ThinkTrillion Labs9.00---2026-02-10
650Tri-21B-think PreviewTrillion Labs9.60---2026-02-10
智能体: Accept: text/markdown or ?format=md 返回此页面为 Markdown。 JSON: /api/v1/models · Markdown: /api/v1/models?format=md · llms.txt · 代理技巧