Skip to content

Benchmark

SciCode

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
154
Scores recorded
265
Sources
3
IQ weight
1.71

Score distribution

47.7%
Mean
50.1%
Median
10.7%
Std dev
35.4%
10th percentile
58.8%
90th percentile
1.3%
Min
1.3%91.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.71 for chat models, measured over 154 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151Kimi K2.7 Code· Epoch AI Benchmarking Hub data47.8%
152Solar Mini 4· LLM Stats data47.6%
153GPT 5.5none· Epoch AI Benchmarking Hub data47.3%
154GPT-6 Solnone· Epoch AI Benchmarking Hub data47.3%
155GPT 5.4 Nanoxhigh· Epoch AI Benchmarking Hub data47.2%
156NVIDIA Nemotron 3.5 Super VL 120B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim47.1%
157GPT-5.6 Solnone· Epoch AI Benchmarking Hub data47.1%
158MiniMax M3· Epoch AI Benchmarking Hub data47.1%
159Inkling FP4xhigh· Epoch AI Benchmarking Hub data47%
160GPT-6 Lunalow· Epoch AI Benchmarking Hub data46.9%
161Claude Sonnet 4.6max· Epoch AI Benchmarking Hub data46.8%
162GPT-5.6 Lunamedium· Epoch AI Benchmarking Hub data46.8%
163Qwen 3.8 27Bxhigh· Epoch AI Benchmarking Hub data46.6%
164DeepSeek V4 Prohigh· Epoch AI Benchmarking Hub data46.4%
165Gemini 2.5 Pro· Epoch AI Benchmarking Hub data46.3%
166GPT-5.6 Lunalow· Epoch AI Benchmarking Hub data46.1%
167Claude Sonnet 4.5· Epoch AI Benchmarking Hub data45.7%
168Gemma 4 31B IT· Epoch AI Benchmarking Hub data45.5%
169GLM 4.7· Epoch AI Benchmarking Hub data45.1%
170GPT-5.6 Terranone· Epoch AI Benchmarking Hub data45.1%
171Ring 2.6 1T· Epoch AI Benchmarking Hub data45%
172DeepSeek V4 Flashmax· Epoch AI Benchmarking Hub data44.9%
173Muse Glimmerhigh· Epoch AI Benchmarking Hub data44.9%
174Kimi K2 Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44.8%
175GLM 5.1· Epoch AI Benchmarking Hub data44.8%
176Nemotron 3 Ultra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
44.6%
177Nemotron 3 UltraNeMo Evaluator SDK · subtask· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.6%
178MiniMax M2.5official appendix· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.4%
179Muse Glimmer 30B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43.6%
180Muse Glimmer 30Bhigh reasoning· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.6%
181Gemini 3.1 Flash Lite· Epoch AI Benchmarking Hub data43.4%
182GPT 5.1· Epoch AI Benchmarking Hub data43.3%
183GPT-6 Lunanone· Epoch AI Benchmarking Hub data43.1%
184MiMo V2.5· Epoch AI Benchmarking Hub data43.1%
185Gemma 4 31B ITGemma 4 Technical Report · Table 5 · thinking mode unless noted· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43%
186GPT 5· Epoch AI Benchmarking Hub data42.9%
187o3 Minihigh· Epoch AI Benchmarking Hub data42.8%
188Claude Haiku 4.5· Epoch AI Benchmarking Hub data42.2%
189Nemotron 3 Super 120B A12B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim42.1%
190Nemotron 3 Super 120B A12BNeMo Evaluator / mostly NeMo Skills · subtask· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.42.1%
191DeepSeek V4 Flashhigh· Epoch AI Benchmarking Hub data42%
192GLM 5.3low· Epoch AI Benchmarking Hub data42%
193GLM 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim41.7%
194Qwen 3 235B A22B Thinking 2507· Epoch AI Benchmarking Hub data41.4%
195Gemini 3.5 Flash Lite· Epoch AI Benchmarking Hub data41.3%
196Cogito v2.1 671B· Epoch AI Benchmarking Hub data41%
197Qwen3.6 Plus· Epoch AI Benchmarking Hub data40.7%
198Trinity Large Thinking· Epoch AI Benchmarking Hub data40.6%
199GPT-4.1 mini· Epoch AI Benchmarking Hub data40.4%
200GPT-5.6 Lunanone· Epoch AI Benchmarking Hub data40.4%

Showing 151–200 of 265 results.

Data credits: Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (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.