Human Leaderboard: overall

Updated Feb. 17, 2025

  • 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=422) Market Score (resolved) (N=49) Market Score (unresolved) (N=27) Market Score (overall) (N=76) Overall Resolved Score (N=471) Overall Score (N=498) Overall Score 95% CI Pairwise p-value comparing to No. 1 (bootstrapped) Pct. more accurate than No. 1 Pct. Imputed
1 ForecastBench Superforecaster median forecast 0.118 0.084 0.058 0.074 0.101 0.096 [0.076, 0.116] 0% 0%
2 ForecastBench Public median forecast 0.153 0.127 0.021 0.089 0.140 0.121 [0.101, 0.141] <0.001 22% 0%
3 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with freeze values) 0.138 0.153 0.023 0.107 0.145 0.122 [0.099, 0.146] <0.001 31% 0%
4 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with news with freeze values) 0.142 0.155 0.035 0.112 0.148 0.127 [0.104, 0.15] <0.001 29% 0%
5 OpenAI GPT-4-Turbo-2024-04-09 (zero shot with freeze values) 0.162 0.136 0.020 0.095 0.149 0.128 [0.105, 0.151] <0.001 32% 0%
6 Anthropic Claude-3-5-Sonnet-20240620 (zero shot with freeze values) 0.145 0.168 0.025 0.117 0.157 0.131 [0.103, 0.159] <0.001 31% 0%
7 OpenAI GPT-4 (zero shot with freeze values) 0.167 0.136 0.024 0.096 0.152 0.132 [0.109, 0.155] <0.001 31% 0%
8 OpenAI GPT-4o (scratchpad with news with freeze values) 0.162 0.136 0.048 0.105 0.149 0.133 [0.113, 0.154] <0.001 25% 0%
9 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad) 0.138 0.179 0.050 0.133 0.158 0.136 [0.113, 0.158] <0.001 28% 0%
10 OpenAI GPT-4o (scratchpad with freeze values) 0.161 0.153 0.040 0.113 0.157 0.137 [0.115, 0.158] <0.001 27% 0%
11 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with news) 0.142 0.173 0.072 0.137 0.157 0.139 [0.117, 0.161] <0.001 26% 0%
12 Anthropic Claude-3-Opus-20240229 (zero shot with freeze values) 0.163 0.162 0.031 0.115 0.163 0.139 [0.114, 0.164] <0.001 23% 0%
13 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 3) 0.158 0.161 0.054 0.123 0.159 0.140 [0.12, 0.161] <0.001 25% 4%
14 OpenAI GPT-4o (scratchpad) 0.161 0.160 0.060 0.125 0.161 0.143 [0.125, 0.16] <0.001 24% 0%
15 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 1) 0.150 0.180 0.054 0.135 0.165 0.143 [0.12, 0.166] <0.001 25% 0%
16 OpenAI GPT-4o (scratchpad with news) 0.162 0.164 0.056 0.126 0.163 0.144 [0.122, 0.165] <0.001 22% 0%
17 Anthropic Claude-3-Opus-20240229 (scratchpad with freeze values) 0.160 0.157 0.077 0.129 0.159 0.144 [0.125, 0.163] <0.001 23% 0%
18 Mistral AI Mistral-Large-Latest (zero shot with freeze values) 0.173 0.164 0.031 0.117 0.168 0.145 [0.121, 0.169] <0.001 23% 0%
19 Google Gemini-1.5-Pro (scratchpad with news with freeze values) 0.163 0.166 0.064 0.130 0.164 0.146 [0.127, 0.165] <0.001 23% 1%
20 Google Gemini-1.5-Pro (scratchpad) 0.162 0.169 0.063 0.131 0.166 0.147 [0.129, 0.164] <0.001 23% 1%
21 Mistral AI Mistral-Large-Latest (scratchpad with freeze values) 0.161 0.167 0.072 0.133 0.164 0.147 [0.129, 0.165] <0.001 22% 0%
22 OpenAI GPT-4-Turbo-2024-04-09 (zero shot) 0.162 0.174 0.069 0.137 0.168 0.149 [0.129, 0.17] <0.001 23% 0%
23 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with freeze values) 0.176 0.155 0.066 0.123 0.166 0.150 [0.126, 0.173] <0.001 26% 0%
24 OpenAI GPT-4 (scratchpad with freeze values) 0.174 0.171 0.046 0.126 0.172 0.150 [0.13, 0.17] <0.001 21% 1%
25 Google Gemini-1.5-Pro (scratchpad with freeze values) 0.162 0.168 0.082 0.138 0.165 0.150 [0.132, 0.169] <0.001 22% 0%
26 Anthropic Claude-3-5-Sonnet-20240620 (zero shot) 0.145 0.194 0.085 0.155 0.170 0.150 [0.123, 0.177] <0.001 24% 0%
27 Google Gemini-1.5-Pro (scratchpad with news) 0.163 0.168 0.094 0.141 0.165 0.152 [0.133, 0.171] <0.001 22% 0%
28 Anthropic Claude-3-Opus-20240229 (superforecaster with news 1) 0.157 0.192 0.067 0.147 0.174 0.152 [0.131, 0.174] <0.001 22% 1%
29 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with news with freeze values) 0.178 0.137 0.112 0.128 0.157 0.153 [0.13, 0.175] <0.001 25% 1%
30 OpenAI GPT-4o (zero shot with freeze values) 0.197 0.158 0.020 0.109 0.178 0.153 [0.128, 0.178] <0.001 26% 0%
31 Meta Llama-3-70b-Chat-Hf (scratchpad with freeze values) 0.191 0.154 0.047 0.116 0.172 0.153 [0.135, 0.171] <0.001 25% 0%
32 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with freeze values) 0.181 0.169 0.052 0.127 0.175 0.154 [0.137, 0.172] <0.001 21% 0%
33 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad) 0.176 0.169 0.066 0.133 0.173 0.154 [0.137, 0.171] <0.001 23% 0%
34 Google Gemini-1.5-Pro (zero shot with freeze values) 0.185 0.173 0.036 0.124 0.179 0.155 [0.127, 0.182] <0.001 25% 0%
35 Anthropic Claude-3-Opus-20240229 (scratchpad) 0.160 0.196 0.064 0.149 0.178 0.155 [0.136, 0.173] <0.001 22% 1%
36 OpenAI GPT-4 (scratchpad) 0.174 0.172 0.074 0.137 0.173 0.155 [0.14, 0.171] <0.001 18% 1%
37 Qwen Qwen1.5-110B-Chat (zero shot with freeze values) 0.197 0.157 0.040 0.115 0.177 0.156 [0.134, 0.179] <0.001 20% 0%
38 Anthropic Claude-2.1 (scratchpad with freeze values) 0.217 0.118 0.055 0.095 0.167 0.156 [0.138, 0.175] <0.001 24% 23%
39 Google Gemini-1.5-Flash (zero shot with freeze values) 0.191 0.173 0.039 0.125 0.182 0.158 [0.13, 0.186] <0.001 23% 0%
40 Mistral AI Mixtral-8x22B-Instruct-V0.1 (zero shot with freeze values) 0.185 0.165 0.069 0.131 0.175 0.158 [0.131, 0.185] <0.001 24% 0%
41 Meta Llama-3-70b-Chat-Hf (zero shot with freeze values) 0.194 0.174 0.034 0.124 0.184 0.159 [0.133, 0.185] <0.001 25% 0%
42 OpenAI GPT-4-Turbo-2024-04-09 (scratchpad with news) 0.178 0.180 0.078 0.144 0.179 0.161 [0.14, 0.182] <0.001 24% 0%
43 ForecastBench Imputed Forecaster 0.250 0.103 0.018 0.073 0.177 0.162 [0.142, 0.181] <0.001 26% 100%
44 Qwen Qwen1.5-110B-Chat (scratchpad with freeze values) 0.183 0.182 0.067 0.141 0.183 0.162 [0.144, 0.18] <0.001 18% 0%
45 Anthropic Claude-2.1 (scratchpad) 0.217 0.121 0.087 0.109 0.169 0.163 [0.143, 0.183] <0.001 22% 24%
46 Qwen Qwen1.5-110B-Chat (scratchpad with news with freeze values) 0.179 0.188 0.074 0.148 0.184 0.163 [0.144, 0.183] <0.001 22% 0%
47 Anthropic Claude-2.1 (zero shot with freeze values) 0.221 0.155 0.021 0.108 0.188 0.164 [0.141, 0.188] <0.001 27% 1%
48 Mistral AI Mistral-Large-Latest (scratchpad) 0.161 0.198 0.113 0.168 0.180 0.165 [0.146, 0.183] <0.001 22% 0%
49 Anthropic Claude-3-5-Sonnet-20240620 (scratchpad with SECOND news) 0.197 0.162 0.079 0.133 0.180 0.165 [0.144, 0.186] <0.001 19% 0%
50 Anthropic Claude-3-Opus-20240229 (zero shot) 0.163 0.220 0.072 0.167 0.192 0.165 [0.14, 0.191] <0.001 18% 0%
51 OpenAI GPT-4o (superforecaster with news 3) 0.198 0.177 0.056 0.134 0.187 0.166 [0.146, 0.186] <0.001 21% 9%
52 OpenAI GPT-4 (zero shot) 0.167 0.197 0.107 0.165 0.182 0.166 [0.147, 0.185] <0.001 22% 0%
53 Google Gemini-1.5-Pro (superforecaster with news 3) 0.183 0.179 0.098 0.150 0.181 0.166 [0.147, 0.186] <0.001 22% 0%
54 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with news with freeze values) 0.201 0.175 0.059 0.133 0.188 0.167 [0.149, 0.186] <0.001 23% 0%
55 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad with news) 0.201 0.170 0.073 0.136 0.186 0.168 [0.152, 0.185] <0.001 22% 0%
56 OpenAI GPT-4o (zero shot) 0.197 0.174 0.078 0.140 0.186 0.169 [0.147, 0.19] <0.001 21% 0%
57 Qwen Qwen1.5-110B-Chat (scratchpad with news) 0.179 0.200 0.088 0.160 0.190 0.170 [0.15, 0.189] <0.001 21% 0%
58 Mistral AI Mixtral-8x22B-Instruct-V0.1 (scratchpad) 0.181 0.192 0.098 0.159 0.187 0.170 [0.152, 0.187] <0.001 20% 0%
59 Qwen Qwen1.5-110B-Chat (scratchpad) 0.183 0.201 0.081 0.158 0.192 0.171 [0.152, 0.189] <0.001 19% 0%
60 Anthropic Claude-3-Opus-20240229 (scratchpad with news with freeze values) 0.183 0.199 0.090 0.161 0.191 0.172 [0.15, 0.193] <0.001 23% 0%
61 Anthropic Claude-3-Opus-20240229 (superforecaster with news 3) 0.177 0.209 0.091 0.167 0.193 0.172 [0.15, 0.194] <0.001 22% 8%
62 Anthropic Claude-3-5-Sonnet-20240620 (superforecaster with news 2) 0.194 0.193 0.073 0.151 0.194 0.172 [0.149, 0.195] <0.001 22% 0%
63 Meta Llama-3-8b-Chat-Hf (zero shot with freeze values) 0.206 0.167 0.090 0.139 0.186 0.173 [0.147, 0.198] <0.001 23% 0%
64 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 3) 0.199 0.180 0.088 0.147 0.190 0.173 [0.151, 0.195] <0.001 20% 13%
65 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad) 0.205 0.162 0.112 0.144 0.183 0.174 [0.154, 0.195] <0.001 27% 15%
66 Anthropic Claude-3-Opus-20240229 (scratchpad with news) 0.183 0.208 0.093 0.167 0.195 0.175 [0.154, 0.196] <0.001 22% 0%
67 Google Gemini-1.5-Flash (scratchpad with freeze values) 0.214 0.169 0.077 0.136 0.192 0.175 [0.153, 0.198] <0.001 18% 0%
68 Mistral AI Mistral-Large-Latest (zero shot) 0.173 0.205 0.131 0.179 0.189 0.176 [0.15, 0.202] <0.001 20% 0%
69 Meta Llama-3-8b-Chat-Hf (scratchpad with freeze values) 0.225 0.165 0.062 0.128 0.195 0.177 [0.161, 0.192] <0.001 23% 0%
70 Anthropic Claude-3-Opus-20240229 (superforecaster with news 2) 0.174 0.210 0.125 0.180 0.192 0.177 [0.152, 0.202] <0.001 22% 0%
71 Qwen Qwen1.5-110B-Chat (superforecaster with news 1) 0.203 0.200 0.068 0.153 0.201 0.178 [0.153, 0.203] <0.001 22% 0%
72 Google Gemini-1.5-Pro (zero shot) 0.185 0.215 0.093 0.171 0.200 0.178 [0.15, 0.206] <0.001 22% 0%
73 Mistral AI Mixtral-8x7B-Instruct-V0.1 (zero shot with freeze values) 0.225 0.187 0.030 0.131 0.206 0.178 [0.148, 0.208] <0.001 28% 0%
74 Mistral AI Mistral-Large-Latest (scratchpad with news with freeze values) 0.201 0.180 0.112 0.156 0.191 0.178 [0.158, 0.199] <0.001 22% 0%
75 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 3) 0.214 0.165 0.104 0.143 0.189 0.179 [0.161, 0.196] <0.001 16% 20%
76 Qwen Qwen1.5-110B-Chat (zero shot) 0.197 0.199 0.093 0.161 0.198 0.179 [0.16, 0.199] <0.001 15% 1%
77 Anthropic Claude-2.1 (scratchpad with news with freeze values) 0.214 0.176 0.091 0.146 0.195 0.180 [0.158, 0.202] <0.001 22% 4%
78 OpenAI GPT-4o (superforecaster with news 1) 0.202 0.204 0.074 0.158 0.203 0.180 [0.153, 0.207] <0.001 25% 0%
79 Google Gemini-1.5-Flash (scratchpad with news with freeze values) 0.220 0.164 0.099 0.141 0.192 0.181 [0.157, 0.204] <0.001 21% 0%
80 OpenAI GPT-4o (scratchpad with SECOND news) 0.236 0.159 0.066 0.126 0.198 0.181 [0.162, 0.2] <0.001 17% 5%
81 Meta Llama-3-8b-Chat-Hf (zero shot) 0.206 0.191 0.094 0.156 0.198 0.181 [0.155, 0.208] <0.001 23% 0%
82 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 1) 0.212 0.190 0.091 0.155 0.201 0.184 [0.158, 0.209] <0.001 21% 0%
83 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.237 0.167 0.065 0.131 0.202 0.184 [0.166, 0.202] <0.001 15% 44%
84 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.237 0.168 0.065 0.131 0.202 0.184 [0.166, 0.202] <0.001 16% 44%
85 ForecastBench LLM Crowd (gpt-4o, claude-3.5-sonnet, gemini-1.5-pro) with news 0.237 0.168 0.064 0.131 0.203 0.184 [0.166, 0.203] <0.001 16% 44%
86 Anthropic Claude-2.1 (scratchpad with news) 0.214 0.181 0.106 0.155 0.198 0.184 [0.161, 0.207] <0.001 21% 9%
87 Google Gemini-1.5-Pro (superforecaster with news 1) 0.201 0.221 0.070 0.168 0.211 0.185 [0.16, 0.209] <0.001 22% 0%
88 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 1) 0.195 0.218 0.098 0.175 0.207 0.185 [0.16, 0.21] <0.001 20% 0%
89 Meta Llama-3-70b-Chat-Hf (zero shot) 0.194 0.218 0.102 0.177 0.206 0.186 [0.162, 0.209] <0.001 21% 0%
90 Mistral AI Mistral-Large-Latest (scratchpad with news) 0.201 0.211 0.102 0.172 0.206 0.187 [0.165, 0.208] <0.001 21% 0%
91 OpenAI GPT-4o (superforecaster with news 2) 0.229 0.171 0.098 0.145 0.200 0.187 [0.162, 0.212] <0.001 23% 1%
92 Meta Llama-3-70b-Chat-Hf (scratchpad) 0.191 0.220 0.118 0.184 0.206 0.187 [0.171, 0.204] <0.001 24% 0%
93 Mistral AI Mixtral-8x22B-Instruct-V0.1 (zero shot) 0.185 0.224 0.130 0.191 0.205 0.188 [0.161, 0.215] <0.001 19% 0%
94 Google Gemini-1.5-Flash (scratchpad) 0.214 0.195 0.103 0.162 0.204 0.188 [0.167, 0.209] <0.001 17% 0%
95 Google Gemini-1.5-Flash (zero shot) 0.191 0.227 0.121 0.189 0.209 0.190 [0.162, 0.217] <0.001 18% 0%
96 Google Gemini-1.5-Flash (scratchpad with news) 0.220 0.182 0.118 0.159 0.201 0.190 [0.168, 0.212] <0.001 20% 0%
97 Mistral AI Mistral-Large-Latest (superforecaster with news 1) 0.212 0.208 0.098 0.169 0.210 0.190 [0.165, 0.216] <0.001 22% 0%
98 Qwen Qwen1.5-110B-Chat (superforecaster with news 3) 0.212 0.213 0.092 0.170 0.213 0.191 [0.171, 0.211] <0.001 21% 7%
99 Mistral AI Mixtral-8x7B-Instruct-V0.1 (zero shot) 0.225 0.190 0.110 0.162 0.208 0.193 [0.165, 0.222] <0.001 21% 0%
100 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 2) 0.251 0.162 0.090 0.137 0.207 0.194 [0.169, 0.219] <0.001 29% 22%
101 Anthropic Claude-2.1 (zero shot) 0.221 0.214 0.090 0.170 0.218 0.196 [0.175, 0.216] <0.001 19% 0%
102 OpenAI GPT-4-Turbo-2024-04-09 (superforecaster with news 2) 0.217 0.212 0.107 0.175 0.215 0.196 [0.171, 0.221] <0.001 25% 2%
103 Meta Llama-3-8b-Chat-Hf (scratchpad) 0.225 0.224 0.072 0.170 0.224 0.197 [0.18, 0.215] <0.001 22% 0%
104 Mistral AI Mixtral-8x22B-Instruct-V0.1 (superforecaster with news 2) 0.235 0.194 0.099 0.160 0.215 0.198 [0.179, 0.217] <0.001 23% 2%
105 Mistral AI Mistral-Large-Latest (superforecaster with news 3) 0.231 0.196 0.109 0.165 0.214 0.198 [0.177, 0.219] <0.001 21% 9%
106 Mistral AI Mistral-Large-Latest (superforecaster with news 2) 0.209 0.225 0.119 0.187 0.217 0.198 [0.173, 0.224] <0.001 21% 1%
107 Anthropic Claude-2.1 (superforecaster with news 3) 0.227 0.195 0.124 0.170 0.211 0.199 [0.176, 0.221] <0.001 23% 6%
108 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 1) 0.248 0.194 0.073 0.151 0.221 0.200 [0.174, 0.226] <0.001 26% 15%
109 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with freeze values) 0.205 0.229 0.150 0.201 0.217 0.203 [0.174, 0.232] <0.001 24% 11%
110 Google Gemini-1.5-Pro (superforecaster with news 2) 0.220 0.221 0.123 0.186 0.220 0.203 [0.174, 0.232] <0.001 21% 0%
111 Qwen Qwen1.5-110B-Chat (superforecaster with news 2) 0.225 0.232 0.096 0.183 0.228 0.204 [0.183, 0.226] <0.001 22% 4%
112 Meta Llama-2-70b-Chat-Hf (zero shot with freeze values) 0.233 0.234 0.072 0.177 0.234 0.205 [0.178, 0.232] <0.001 25% 1%
113 Google Gemini-1.5-Flash (superforecaster with news 3) 0.236 0.212 0.108 0.175 0.224 0.206 [0.182, 0.23] <0.001 19% 11%
114 Meta Llama-2-70b-Chat-Hf (scratchpad with freeze values) 0.226 0.234 0.098 0.186 0.230 0.206 [0.19, 0.222] <0.001 22% 0%
115 Google Gemini-1.5-Flash (superforecaster with news 2) 0.228 0.216 0.139 0.188 0.222 0.208 [0.182, 0.234] <0.001 20% 0%
116 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with news with freeze values) 0.295 0.129 0.116 0.124 0.212 0.210 [0.19, 0.229] <0.001 23% 13%
117 Anthropic Claude-3-Haiku-20240307 (scratchpad with freeze values) 0.238 0.231 0.094 0.182 0.234 0.210 [0.191, 0.229] <0.001 22% 0%
118 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 2) 0.233 0.221 0.127 0.187 0.227 0.210 [0.191, 0.229] <0.001 20% 0%
119 Mistral AI Mixtral-8x7B-Instruct-V0.1 (superforecaster with news 3) 0.250 0.184 0.152 0.173 0.217 0.211 [0.189, 0.233] <0.001 23% 14%
120 Google Gemini-1.5-Flash (superforecaster with news 1) 0.231 0.245 0.103 0.194 0.238 0.213 [0.185, 0.241] <0.001 22% 1%
121 Anthropic Claude-3-Haiku-20240307 (scratchpad) 0.238 0.228 0.116 0.188 0.233 0.213 [0.195, 0.231] <0.001 22% 0%
122 Anthropic Claude-3-Haiku-20240307 (zero shot with freeze values) 0.301 0.161 0.070 0.129 0.231 0.215 [0.197, 0.233] <0.001 19% 1%
123 Meta Llama-2-70b-Chat-Hf (scratchpad) 0.226 0.253 0.141 0.213 0.240 0.220 [0.201, 0.239] <0.001 22% 1%
124 Anthropic Claude-2.1 (superforecaster with news 2) 0.241 0.235 0.137 0.200 0.238 0.221 [0.195, 0.246] <0.001 23% 11%
125 Anthropic Claude-2.1 (superforecaster with news 1) 0.260 0.230 0.096 0.183 0.245 0.221 [0.197, 0.246] <0.001 23% 4%
126 OpenAI GPT-3.5-Turbo-0125 (scratchpad with freeze values) 0.258 0.241 0.105 0.192 0.249 0.225 [0.206, 0.245] <0.001 22% 0%
127 Anthropic Claude-3-Haiku-20240307 (scratchpad with news with freeze values) 0.277 0.215 0.105 0.176 0.246 0.226 [0.209, 0.244] <0.001 21% 0%
128 Anthropic Claude-3-Haiku-20240307 (scratchpad with news) 0.277 0.212 0.114 0.177 0.244 0.227 [0.21, 0.244] <0.001 21% 0%
129 Mistral AI Mixtral-8x7B-Instruct-V0.1 (scratchpad with news) 0.295 0.192 0.103 0.161 0.244 0.228 [0.204, 0.252] <0.001 22% 14%
130 ForecastBench Always 0.5 0.250 0.250 0.132 0.208 0.250 0.229 [0.22, 0.238] <0.001 16% 0%
131 OpenAI GPT-3.5-Turbo-0125 (scratchpad) 0.258 0.249 0.127 0.206 0.254 0.232 [0.212, 0.252] <0.001 20% 0%
132 Anthropic Claude-3-Haiku-20240307 (zero shot) 0.301 0.204 0.103 0.168 0.253 0.235 [0.216, 0.253] <0.001 18% 1%
133 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 3) 0.266 0.244 0.148 0.210 0.255 0.238 [0.219, 0.257] <0.001 18% 26%
134 Meta Llama-2-70b-Chat-Hf (zero shot) 0.233 0.322 0.127 0.252 0.277 0.243 [0.215, 0.27] <0.001 23% 1%
135 Anthropic Claude-3-Haiku-20240307 (superforecaster with news 1) 0.286 0.295 0.113 0.230 0.290 0.258 [0.232, 0.284] <0.001 22% 0%
136 OpenAI GPT-3.5-Turbo-0125 (zero shot with freeze values) 0.422 0.159 0.073 0.128 0.290 0.275 [0.247, 0.304] <0.001 29% 0%
137 ForecastBench Always 0 0.329 0.286 0.208 0.258 0.308 0.294 [0.242, 0.345] <0.001 41% 0%
138 ForecastBench Random Uniform 0.357 0.302 0.194 0.264 0.329 0.310 [0.274, 0.346] <0.001 22% 0%
139 OpenAI GPT-3.5-Turbo-0125 (zero shot) 0.422 0.256 0.171 0.226 0.339 0.324 [0.293, 0.355] <0.001 21% 0%
140 ForecastBench Always 1 0.671 0.714 0.556 0.658 0.692 0.664 [0.61, 0.719] <0.001 26% 0%