Leaderboard: 30 day

  • Ranking: The position of the model in the leaderboard as ordered by Overall Score
  • Organization: The group responsible for the model or forecasts
  • Model: The LLM model & prompt info or the human group and forecast aggregation method
    • zero shot: used a zero-shot prompt
    • scratchpad: used a scratchpad prompt with instructions that outline a procedure the model should use to reason about the question
    • with freeze values: means that, for questions from market sources, the prompt was supplemented with the aggregate human forecast from the relevant platform on the day the question set was generated
    • with news: means that the prompt was supplemented with relevant news summaries obtained through an automated process
  • Dataset Score: The average Brier score across all questions sourced from datasets
  • Market Score (resolved): The average Brier score across all resolved questions sourced from prediction markets and forecast aggregation platforms
  • Market Score (unresolved): The average Brier score across all unresolved questions sourced from prediction markets and forecast aggregation platforms
  • Market Score (overall): The average Brier score across all questions sourced from prediction markets and forecast aggregation platforms
  • Overall Resolved Score: The average of the Dataset Score and the Market Score (resolved) columns
  • Overall Score: The average of the Dataset Score and the Market Score (overall) columns
  • Overall Score 95% CI: The 95% confidence interval for the Overall Score
  • Pairwise p-value comparing to No. 1 (bootstrapped): The p-value calculated by bootstrapping the differences in overall score between each model and the best forecaster (the group with rank 1) under the null hypothesis that there's no difference.
  • Pct. more accurate than No. 1: The percent of questions where this forecaster had a better overall score than the best forecaster (with rank 1)
  • Pct. imputed: The percent of questions for which this forecaster did not provide a forecast and hence had a forecast value imputed (0.5 for dataset questions and the aggregate human forecast on the forecast due date for questions sourced from prediction markets or forecast aggregation platforms)
Ranking Organization Model Dataset Score (N=1,388) Market Score (resolved) (N=85) Market Score (unresolved) (N=827) Market Score (overall) (N=912) Overall Resolved Score (N=1,473) Overall Score (N=2,300) Overall Score 95% CI Pairwise p-value comparing to No. 1 (bootstrapped) Pct. more accurate than No. 1 Pct. Imputed
1 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with freeze values) 0.174 0.110 0.041 0.047 0.142 0.111 [0.104, 0.117] 0% 0%
2 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with freeze values) 0.177 0.109 0.045 0.051 0.143 0.114 [0.107, 0.12] 0.183 43% 0%
3 OpenAI GPT-4o (scratchpad with freeze values) 0.200 0.095 0.033 0.039 0.147 0.119 [0.113, 0.125] 0.003 42% 0%
4 Google Gemini-1.5-Pro (scratchpad with freeze values) 0.165 0.135 0.069 0.075 0.150 0.120 [0.113, 0.127] 0.009 34% 0%
5 OpenAI GPT-4o (scratchpad with news with freeze values) 0.194 0.105 0.044 0.050 0.150 0.122 [0.116, 0.128] <0.001 39% 0%
6 Google Gemini-1.5-Pro (scratchpad with news with freeze values) 0.167 0.143 0.076 0.082 0.155 0.125 [0.118, 0.131] <0.001 34% 0%
7 Anthropic Claude-3-Opus-20240229 (zero shot with freeze values) 0.190 0.094 0.058 0.061 0.142 0.126 [0.118, 0.133] <0.001 41% 0%
8 Qwen Qwen1.5-110B-Chat (scratchpad with freeze values) 0.179 0.126 0.073 0.078 0.153 0.129 [0.123, 0.135] <0.001 31% 0%
9 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad) 0.177 0.162 0.074 0.083 0.169 0.130 [0.123, 0.136] <0.001 31% 0%
10 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad) 0.174 0.177 0.077 0.087 0.176 0.130 [0.124, 0.137] <0.001 9% 0%
11 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with news with freeze values) 0.194 0.118 0.062 0.067 0.156 0.131 [0.124, 0.137] <0.001 31% 0%
12 OpenAI GPT-4-Turbo-2024-04-09 (zero shot with freeze values) 0.206 0.106 0.050 0.055 0.156 0.131 [0.123, 0.138] <0.001 42% 0%
13 Anthropic Claude-3-5-Sonnet-20240620 (zero shot with freeze values) 0.199 0.130 0.056 0.063 0.164 0.131 [0.123, 0.138] <0.001 43% 0%
14 Google Gemini-1.5-Pro (scratchpad) 0.165 0.167 0.090 0.097 0.166 0.131 [0.124, 0.138] <0.001 32% 0%
15 Google Gemini-1.5-Pro (scratchpad with news) 0.167 0.167 0.089 0.097 0.167 0.132 [0.125, 0.139] <0.001 32% 0%
16 OpenAI GPT-4 (scratchpad with freeze values) 0.199 0.135 0.061 0.067 0.167 0.133 [0.127, 0.139] <0.001 35% 0%
17 Google Gemini-1.5-Pro (zero shot with freeze values) 0.219 0.084 0.048 0.052 0.151 0.135 [0.128, 0.142] <0.001 39% 17%
18 ForecastBench Imputed Forecaster 0.250 0.043 0.019 0.021 0.146 0.136 [0.133, 0.138] <0.001 47% 100%
19 Meta Llama-3-70b-Chat-Hf (zero shot with freeze values) 0.205 0.116 0.064 0.069 0.160 0.137 [0.13, 0.144] <0.001 33% 0%
20 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with news with freeze values) 0.215 0.103 0.056 0.061 0.159 0.138 [0.132, 0.144] <0.001 34% 0%
21 OpenAI GPT-4o (scratchpad with news) 0.194 0.133 0.077 0.082 0.163 0.138 [0.132, 0.145] <0.001 30% 0%
22 Google Gemini-1.5-Flash (scratchpad with freeze values) 0.188 0.138 0.083 0.088 0.163 0.138 [0.131, 0.146] <0.001 31% 0%
23 OpenAI GPT-4 (zero shot with freeze values) 0.223 0.100 0.053 0.057 0.162 0.140 [0.133, 0.147] <0.001 37% 0%
24 OpenAI GPT-4o (scratchpad) 0.200 0.196 0.069 0.081 0.198 0.140 [0.134, 0.147] <0.001 30% 0%
25 Qwen Qwen1.5-110B-Chat (scratchpad) 0.179 0.149 0.097 0.102 0.164 0.141 [0.135, 0.146] <0.001 29% 0%
26 Mistral AI Mistral-Large-Latest (scratchpad with freeze values) 0.202 0.120 0.078 0.082 0.161 0.142 [0.136, 0.148] <0.001 26% 0%
27 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with news) 0.194 0.138 0.087 0.091 0.166 0.143 [0.136, 0.149] <0.001 29% 0%
28 Anthropic Claude-3-Opus-20240229 (scratchpad with freeze values) 0.207 0.125 0.075 0.080 0.166 0.143 [0.137, 0.15] <0.001 27% 0%
29 Google Gemini-1.5-Pro (zero shot) 0.219 0.133 0.063 0.069 0.176 0.144 [0.137, 0.151] <0.001 36% 17%
30 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.250 0.087 0.034 0.039 0.168 0.145 [0.142, 0.147] <0.001 38% 83%
31 Google Gemini-1.5-Pro (superforecaster with news 3) 0.191 0.172 0.092 0.099 0.181 0.145 [0.138, 0.152] <0.001 31% 0%
32 Meta Llama-3-70b-Chat-Hf (scratchpad with freeze values) 0.215 0.130 0.070 0.075 0.172 0.145 [0.14, 0.151] <0.001 25% 0%
33 OpenAI GPT-4 (scratchpad) 0.199 0.136 0.087 0.091 0.168 0.145 [0.14, 0.151] <0.001 28% 0%
34 OpenAI GPT-4-Turbo-2024-04-09 (zero shot) 0.206 0.149 0.079 0.086 0.178 0.146 [0.138, 0.154] <0.001 31% 0%
35 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 3) 0.196 0.142 0.091 0.096 0.169 0.146 [0.139, 0.153] <0.001 29% 2%
36 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.250 0.086 0.037 0.042 0.168 0.146 [0.143, 0.149] <0.001 38% 83%
37 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.250 0.088 0.037 0.042 0.169 0.146 [0.143, 0.149] <0.001 38% 83%
38 Google Gemini-1.5-Flash (scratchpad) 0.188 0.163 0.100 0.106 0.175 0.147 [0.14, 0.154] <0.001 28% 0%
39 Anthropic Claude-3-Opus-20240229 (zero shot) 0.190 0.158 0.099 0.104 0.174 0.147 [0.14, 0.155] <0.001 33% 0%
40 Mistral AI Mistral-Large-Latest (zero shot with freeze values) 0.206 0.148 0.085 0.091 0.177 0.148 [0.14, 0.156] <0.001 31% 0%
41 Google Gemini-1.5-Flash (zero shot with freeze values) 0.228 0.125 0.063 0.068 0.176 0.148 [0.14, 0.157] <0.001 39% 0%
42 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 3) 0.212 0.127 0.082 0.086 0.170 0.149 [0.143, 0.155] <0.001 27% 11%
43 Qwen Qwen1.5-110B-Chat (scratchpad with news with freeze values) 0.209 0.118 0.087 0.090 0.163 0.149 [0.143, 0.156] <0.001 26% 0%
44 Anthropic Claude-3-5-Sonnet-20240620 (zero shot) 0.199 0.177 0.092 0.100 0.188 0.149 [0.141, 0.157] <0.001 35% 1%
45 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with news) 0.215 0.125 0.080 0.085 0.170 0.150 [0.144, 0.156] <0.001 26% 0%
46 Meta Llama-3-70b-Chat-Hf (zero shot) 0.205 0.141 0.090 0.095 0.173 0.150 [0.143, 0.156] <0.001 28% 0%
47 Anthropic Claude-2.1 (scratchpad) 0.233 0.109 0.069 0.072 0.171 0.152 [0.147, 0.158] <0.001 39% 22%
48 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with freeze values) 0.215 0.141 0.086 0.091 0.178 0.153 [0.147, 0.16] <0.001 29% 0%
49 OpenAI GPT-4o (superforecaster with news 3) 0.214 0.125 0.089 0.093 0.170 0.154 [0.148, 0.159] <0.001 27% 6%
50 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 1) 0.214 0.138 0.092 0.096 0.176 0.155 [0.149, 0.161] <0.001 30% 23%
51 OpenAI GPT-4o (superforecaster with news 1) 0.217 0.136 0.089 0.094 0.176 0.155 [0.147, 0.163] <0.001 30% 0%
52 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 1) 0.216 0.196 0.084 0.095 0.206 0.155 [0.148, 0.162] <0.001 29% 0%
53 Anthropic Claude-2.1 (scratchpad with freeze values) 0.233 0.148 0.075 0.081 0.190 0.157 [0.151, 0.163] <0.001 37% 17%
54 Anthropic Claude-3-Opus-20240229 (scratchpad) 0.207 0.153 0.103 0.107 0.180 0.157 [0.151, 0.164] <0.001 26% 0%
55 Qwen Qwen1.5-110B-Chat (zero shot with freeze values) 0.229 0.129 0.082 0.086 0.179 0.157 [0.15, 0.165] <0.001 31% 0%
56 Google Gemini-1.5-Flash (scratchpad with news with freeze values) 0.217 0.150 0.094 0.099 0.184 0.158 [0.151, 0.166] <0.001 27% 0%
57 OpenAI GPT-4o (zero shot with freeze values) 0.233 0.117 0.081 0.084 0.175 0.159 [0.15, 0.167] <0.001 36% 0%
58 Meta Llama-3-8b-Chat-Hf (zero shot with freeze values) 0.220 0.163 0.093 0.100 0.191 0.160 [0.152, 0.167] <0.001 42% 0%
59 Mistral AI Mistral-Large-Latest (scratchpad) 0.202 0.184 0.111 0.118 0.193 0.160 [0.154, 0.166] <0.001 24% 0%
60 Qwen Qwen1.5-110B-Chat (scratchpad with news) 0.209 0.148 0.108 0.112 0.178 0.160 [0.154, 0.166] <0.001 25% 0%
61 OpenAI GPT-4o (zero shot) 0.233 0.183 0.079 0.089 0.208 0.161 [0.153, 0.168] <0.001 29% 3%
62 Meta Llama-3-70b-Chat-Hf (scratchpad) 0.215 0.182 0.107 0.114 0.199 0.165 [0.159, 0.17] <0.001 23% 0%
63 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad) 0.215 0.195 0.107 0.115 0.205 0.165 [0.159, 0.171] <0.001 26% 0%
64 Anthropic Claude-2.1 (scratchpad with news) 0.240 0.179 0.081 0.090 0.210 0.165 [0.159, 0.172] <0.001 35% 17%
65 Qwen Qwen1.5-110B-Chat (superforecaster with news 1) 0.214 0.231 0.106 0.118 0.223 0.166 [0.159, 0.172] <0.001 31% 21%
66 Mistral AI Mixtral-8x22B-Instruct-V0.1 (zero shot with freeze values) 0.219 0.174 0.110 0.116 0.196 0.167 [0.159, 0.176] <0.001 28% 0%
67 Google Gemini-1.5-Pro (superforecaster with news 1) 0.221 0.191 0.106 0.114 0.206 0.167 [0.16, 0.175] <0.001 29% 5%
68 Anthropic Claude-2.1 (scratchpad with news with freeze values) 0.240 0.158 0.089 0.095 0.199 0.168 [0.161, 0.174] <0.001 34% 14%
69 Anthropic Claude-3-Opus-20240229 (scratchpad with news with freeze values) 0.227 0.142 0.105 0.108 0.185 0.168 [0.161, 0.175] <0.001 23% 0%
70 Mistral AI Mixtral-8x22B-Instruct-V0.1 (zero shot) 0.219 0.153 0.113 0.117 0.186 0.168 [0.16, 0.176] <0.001 25% 0%
71 Qwen Qwen1.5-110B-Chat (superforecaster with news 3) 0.222 0.164 0.109 0.114 0.193 0.168 [0.162, 0.174] <0.001 24% 4%
72 OpenAI GPT-4 (zero shot) 0.223 0.146 0.110 0.113 0.184 0.168 [0.161, 0.175] <0.001 25% 0%
73 Google Gemini-1.5-Flash (scratchpad with news) 0.217 0.176 0.115 0.121 0.197 0.169 [0.162, 0.176] <0.001 25% 0%
74 Mistral AI Mistral-Large-Latest (zero shot) 0.206 0.150 0.131 0.133 0.178 0.169 [0.161, 0.177] <0.001 27% 0%
75 Google Gemini-1.5-Flash (superforecaster with news 2) 0.239 0.131 0.097 0.100 0.185 0.169 [0.162, 0.177] <0.001 27% 12%
76 Anthropic Claude-3-Opus-20240229 (superforecaster with news 1) 0.221 0.180 0.115 0.121 0.200 0.171 [0.164, 0.178] <0.001 26% 12%
77 Qwen Qwen1.5-110B-Chat (zero shot) 0.229 0.158 0.113 0.117 0.193 0.173 [0.166, 0.179] <0.001 26% 0%
78 Meta Llama-3-8b-Chat-Hf (zero shot) 0.220 0.227 0.119 0.129 0.223 0.175 [0.166, 0.183] <0.001 40% 0%
79 Anthropic Claude-3-Opus-20240229 (superforecaster with news 3) 0.222 0.165 0.126 0.130 0.193 0.176 [0.169, 0.183] <0.001 25% 6%
80 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with news with freeze values) 0.235 0.179 0.111 0.117 0.207 0.176 [0.169, 0.182] <0.001 25% 0%
81 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 1) 0.236 0.156 0.112 0.116 0.196 0.176 [0.169, 0.183] <0.001 29% 26%
82 Anthropic Claude-3-Opus-20240229 (scratchpad with news) 0.227 0.169 0.120 0.125 0.198 0.176 [0.169, 0.183] <0.001 23% 0%
83 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 2) 0.225 0.176 0.125 0.130 0.200 0.177 [0.169, 0.185] <0.001 26% 0%
84 Anthropic Claude-3-Opus-20240229 (superforecaster with news 2) 0.229 0.162 0.123 0.127 0.196 0.178 [0.17, 0.185] <0.001 25% 1%
85 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 3) 0.236 0.176 0.115 0.121 0.206 0.178 [0.173, 0.184] <0.001 25% 15%
86 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with news) 0.235 0.206 0.116 0.125 0.220 0.180 [0.173, 0.186] <0.001 24% 0%
87 Google Gemini-1.5-Flash (superforecaster with news 3) 0.230 0.203 0.123 0.130 0.216 0.180 [0.174, 0.187] <0.001 24% 12%
88 Anthropic Claude-2.1 (superforecaster with news 2) 0.270 0.164 0.086 0.093 0.217 0.181 [0.173, 0.19] <0.001 36% 24%
89 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with freeze values) 0.253 0.143 0.108 0.111 0.198 0.182 [0.174, 0.189] <0.001 28% 10%
90 Mistral AI Mixtral-8x7B-Instruct-V0.1 (zero shot with freeze values) 0.266 0.143 0.095 0.099 0.204 0.183 [0.173, 0.193] <0.001 37% 0%
91 Google Gemini-1.5-Pro (superforecaster with news 2) 0.245 0.206 0.112 0.121 0.225 0.183 [0.174, 0.192] <0.001 28% 0%
92 Google Gemini-1.5-Flash (zero shot) 0.228 0.201 0.132 0.138 0.214 0.183 [0.174, 0.192] <0.001 30% 0%
93 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad) 0.253 0.110 0.115 0.115 0.182 0.184 [0.177, 0.191] <0.001 27% 11%
94 Anthropic Claude-2.1 (zero shot with freeze values) 0.245 0.187 0.116 0.123 0.216 0.184 [0.176, 0.192] <0.001 32% 0%
95 OpenAI GPT-4o (superforecaster with news 2) 0.264 0.161 0.102 0.108 0.212 0.186 [0.178, 0.194] <0.001 27% 5%
96 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 3) 0.274 0.164 0.096 0.102 0.219 0.188 [0.181, 0.195] <0.001 30% 18%
97 Qwen Qwen1.5-110B-Chat (superforecaster with news 2) 0.242 0.193 0.128 0.134 0.218 0.188 [0.181, 0.195] <0.001 23% 5%
98 Anthropic Claude-2.1 (superforecaster with news 1) 0.268 0.194 0.102 0.110 0.231 0.189 [0.182, 0.197] <0.001 33% 25%
99 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 2) 0.250 0.176 0.126 0.131 0.213 0.190 [0.184, 0.197] <0.001 23% 2%
100 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 2) 0.256 0.160 0.123 0.127 0.208 0.191 [0.183, 0.2] <0.001 26% 9%
101 Mistral AI Mistral-Large-Latest (scratchpad with news with freeze values) 0.254 0.198 0.126 0.132 0.226 0.193 [0.186, 0.2] <0.001 20% 0%
102 Mistral AI Mistral-Large-Latest (superforecaster with news 1) 0.251 0.199 0.131 0.138 0.225 0.194 [0.187, 0.201] <0.001 27% 35%
103 Mistral AI Mistral-Large-Latest (superforecaster with news 2) 0.243 0.209 0.146 0.152 0.226 0.197 [0.189, 0.205] <0.001 26% 13%
104 Mistral AI Mistral-Large-Latest (scratchpad with news) 0.254 0.187 0.137 0.142 0.221 0.198 [0.191, 0.205] <0.001 20% 0%
105 Anthropic Claude-2.1 (zero shot) 0.245 0.239 0.147 0.155 0.242 0.200 [0.194, 0.206] <0.001 25% 0%
106 Anthropic Claude-2.1 (superforecaster with news 3) 0.264 0.238 0.131 0.141 0.251 0.202 [0.195, 0.21] <0.001 26% 10%
107 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 1) 0.300 0.163 0.099 0.105 0.232 0.203 [0.195, 0.211] <0.001 32% 18%
108 Mistral AI Mixtral-8x7B-Instruct-V0.1 (zero shot) 0.266 0.168 0.138 0.141 0.217 0.203 [0.193, 0.214] <0.001 32% 0%
109 Meta Llama-3-8b-Chat-Hf (scratchpad with freeze values) 0.270 0.198 0.132 0.138 0.234 0.204 [0.197, 0.211] <0.001 22% 0%
110 Google Gemini-1.5-Flash (superforecaster with news 1) 0.246 0.251 0.153 0.162 0.248 0.204 [0.196, 0.212] <0.001 24% 24%
111 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with news) 0.294 0.168 0.111 0.117 0.231 0.205 [0.198, 0.213] <0.001 28% 15%
112 ForecastBench Always 0.5 0.250 0.250 0.151 0.160 0.250 0.205 [0.202, 0.208] <0.001 25% 0%
113 Meta Llama-2-70b-Chat-Hf (scratchpad with freeze values) 0.265 0.214 0.139 0.146 0.239 0.205 [0.199, 0.211] <0.001 22% 0%
114 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with news with freeze values) 0.294 0.175 0.112 0.118 0.234 0.206 [0.198, 0.213] <0.001 27% 15%
115 Anthropic Claude-3-Haiku-20240307 (zero shot with freeze values) 0.281 0.152 0.131 0.133 0.216 0.207 [0.199, 0.214] <0.001 24% 0%
116 Meta Llama-2-70b-Chat-Hf (scratchpad) 0.265 0.226 0.143 0.150 0.245 0.207 [0.202, 0.213] <0.001 22% 0%
117 Mistral AI Mistral-Large-Latest (superforecaster with news 3) 0.274 0.213 0.137 0.144 0.244 0.209 [0.202, 0.216] <0.001 21% 5%
118 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 2) 0.293 0.213 0.124 0.132 0.253 0.212 [0.204, 0.221] <0.001 30% 19%
119 Anthropic Claude-3-Haiku-20240307 (scratchpad with freeze values) 0.277 0.187 0.145 0.149 0.232 0.213 [0.206, 0.219] <0.001 20% 0%
120 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 2) 0.265 0.206 0.166 0.170 0.235 0.217 [0.211, 0.224] <0.001 22% 7%
121 Meta Llama-3-8b-Chat-Hf (scratchpad) 0.270 0.227 0.161 0.167 0.248 0.218 [0.211, 0.225] <0.001 21% 0%
122 Anthropic Claude-3-Haiku-20240307 (zero shot) 0.281 0.204 0.151 0.156 0.243 0.218 [0.212, 0.225] <0.001 23% 0%
123 ForecastBench Always 0 0.268 0.282 0.163 0.174 0.275 0.221 [0.206, 0.236] <0.001 58% 0%
124 Anthropic Claude-3-Haiku-20240307 (scratchpad) 0.277 0.276 0.162 0.173 0.276 0.225 [0.219, 0.231] <0.001 19% 0%
125 Anthropic Claude-3-Haiku-20240307 (scratchpad with news with freeze values) 0.304 0.191 0.156 0.159 0.248 0.232 [0.225, 0.238] <0.001 19% 0%
126 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 3) 0.283 0.257 0.181 0.188 0.270 0.235 [0.229, 0.242] <0.001 22% 21%
127 Anthropic Claude-3-Haiku-20240307 (scratchpad with news) 0.304 0.214 0.168 0.173 0.259 0.238 [0.232, 0.245] <0.001 19% 1%
128 OpenAI GPT-3.5-Turbo-0125 (scratchpad with freeze values) 0.299 0.270 0.175 0.184 0.284 0.241 [0.235, 0.248] <0.001 19% 0%
129 OpenAI GPT-3.5-Turbo-0125 (scratchpad) 0.299 0.271 0.181 0.189 0.285 0.244 [0.237, 0.25] <0.001 19% 0%
130 Meta Llama-2-70b-Chat-Hf (zero shot with freeze values) 0.304 0.284 0.185 0.194 0.294 0.249 [0.24, 0.259] <0.001 26% 0%
131 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 1) 0.296 0.287 0.210 0.218 0.292 0.257 [0.249, 0.265] <0.001 21% 8%
132 Meta Llama-2-70b-Chat-Hf (zero shot) 0.304 0.308 0.212 0.221 0.306 0.263 [0.255, 0.271] <0.001 24% 1%
133 ForecastBench Random Uniform 0.340 0.369 0.221 0.235 0.354 0.287 [0.276, 0.299] <0.001 27% 0%
134 OpenAI GPT-3.5-Turbo-0125 (zero shot with freeze values) 0.450 0.172 0.172 0.172 0.311 0.311 [0.3, 0.322] <0.001 25% 0%
135 OpenAI GPT-3.5-Turbo-0125 (zero shot) 0.450 0.297 0.245 0.250 0.373 0.350 [0.339, 0.361] <0.001 19% 0%
136 ForecastBench Always 1 0.732 0.718 0.639 0.647 0.725 0.689 [0.673, 0.706] <0.001 20% 0%