Skip to content

Benchmark

GSM8K

Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.

Models and systems measured
104
Scores recorded
104
Sources
5
IQ weight
0.30

Score distribution

74%
Mean
87.6%
Median
28%
Std dev
25.2%
10th percentile
95.1%
90th percentile
0%
Min
0%99.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.30 for chat models, measured over 97 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1MiMo V2.5 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
99.6%
2Kimi K2 Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim97.3%
3o1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim97.1%
4GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim97%
5Llama 3.1 405B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim96.8%
6Llama 3.1 405B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim96.8%
7Claude 3.5 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim96.4%
8Gemma 3 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.9%
9Qwen2.5 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.9%
10Qwen2.5 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.8%
11DeepSeek V2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.1%
12Llama 3.1 70B Instruct8-shot · chain of thought· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.95.1%
13Claude 3 Opus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95%
14Nova Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.8%
15Qwen 2.5 14B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.8%
16DeepSeek Coder V2 Instruct· Epoch AI Benchmarking Hub data94.5%
17Nova Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.5%
18Gemma 3 12B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.4%
19Qwen 3 235B A22B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.4%
20qwen2.5-coder-14b-instruct· Epoch AI Benchmarking Hub data94.2%
21Llama 3.2 90B Vision Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim93.1%
22Mistral Large 2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93%
23Qwen 2.5 Coder 32B Instruct· Epoch AI Benchmarking Hub data
Values from 2 sources
93%
24DeepSeek V4 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim92.6%
25Granite 4.1 8B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim92.5%
26Claude 3 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.3%
27Nova Micro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.3%
28Qwen2.5 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.6%
29Llama 3.1 Nemotron 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.4%
30GPT-4o mini· Epoch AI Benchmarking Hub data91.3%
31Qwen2 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.1%
32Qwen 2.5 Coder 32B· Epoch AI Benchmarking Hub data91.1%
33Phi-3 Medium 4K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim91%
34Gemini 1.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.8%
35GPT 4· Epoch AI Benchmarking Hub data90%
36Qwen 2 (72B)· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
89.5%
37DeepSeek V3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
89.3%
38Gemma 3 4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.2%
39Claude 3 Haiku· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.9%
40phi-3.5-moe-instruct· Epoch AI Benchmarking Hub data
Values from 2 sources
88.7%
41Qwen2.5-Omni 7B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
88.7%
42qwen2p5-coder-14b· Epoch AI Benchmarking Hub data88.7%
43Phi 4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.6%
44deepseek-coder-v2-lite-instruct· Epoch AI Benchmarking Hub data87.6%
45Jamba 1.5 Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87%
46Claude Instant 1.2· Epoch AI Benchmarking Hub data86.7%
47Qwen2.5 Coder 7B Instruct· Epoch AI Benchmarking Hub data
Values from 2 sources
86.7%
48Gemini 1.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86.2%
49deepseek-coder-v2-base· Epoch AI Benchmarking Hub data85.8%
50Phi-3 Mini 4K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim85.7%

Showing 1–50 of 98 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
DeepSeek V4.1 Flash BaseEM · 8-shot· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93%
MiMo V2 Flash Baseofficial Base table · 8-shot· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.92.3%
Qwen3.5-397B-A17B-BaseQwen3.5 release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93.7%
Qwen3.7-Plus-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93%
Qwen3.8-27B-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93.2%
Qwen3.8-Flash-Next-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93.3%

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · OpenEvals/leaderboard-data, MIT (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.