PUBLISHED BENCHMARK RESULTS
Aider Code Editing
Historical code editing runs using Aider. Task counts and editing formats vary between runs.
These are publisher-reported model and agent runs. Agent version, reasoning effort or editing format, environment and source dates may differ. Task counts do not prove identical evaluation conditions. Compare setups before drawing conclusions.
Reported costs are evaluation totals in USD, not current API prices. Missing results are not scores of zero. No score is a universal recommendation.
Historical coverage: this source contains older Aider evaluations. Refreshing the source does not make the underlying results current.
Benchmark publisher ↗ · Apache-2.0 ↗ · Full methodology and provenance
133 tasks
- Filtered rank 1
claude-3-5-sonnet-20241022
Source run ↗diff84.2%Compare - Filtered rank 2diff84.2%Compare
- Filtered rank 3
gemini-exp-1206 (whole)
Source run ↗whole80.5%Compare - Filtered rank 4
o1-preview
Source run ↗diff79.7%Compare - Filtered rank 5
claude-3.5-sonnet-20240620
Source run ↗diff77.4%Compare - Filtered rank 6
claude-3-5-haiku-20241022
Source run ↗diff75.2%Compare - Filtered rank 7
DeepSeek Coder V2 0724
Source run ↗diff72.9%Compare - Filtered rank 8
gpt-4o-2024-05-13
Source run ↗diff72.9%Compare - Filtered rank 9
ollama/qwen2.5-coder:32b
Source run ↗whole72.9%Compare - Filtered rank 10
DeepSeek V2.5
Source run ↗diff72.2%Compare - Filtered rank 11
DeepSeek-V2.5-1210
Source run ↗diff72.2%Compare - Filtered rank 12
openai/chatgpt-4o-latest
Source run ↗diff72.2%Compare - Filtered rank 13
Qwen2.5-Coder-32B-Instruct
Source run ↗diff71.4%Compare - Filtered rank 14
gpt-4o-2024-08-06
Source run ↗diff71.4%Compare - Filtered rank 15
gpt-4o-2024-11-20
Source run ↗diff71.4%Compare - Filtered rank 16
o1-mini (whole)
Source run ↗whole70.7%Compare - Filtered rank 17
DeepSeek Chat V2 0628
Source run ↗diff69.9%Compare - Filtered rank 18
gemini-2.0-flash-exp
Source run ↗diff69.9%Compare - Filtered rank 19
Qwen2.5-Coder-14B-Instruct
Source run ↗whole69.2%Compare - Filtered rank 20
gemini-exp-1206 (diff)
Source run ↗diff69.2%Compare - Filtered rank 21
claude-3-opus-20240229
Source run ↗diff68.4%Compare - Filtered rank 22
gpt-4-0613
Source run ↗diff67.7%Compare - Filtered rank 23
Dracarys2-72B-Instruct
Source run ↗whole66.9%Compare - Filtered rank 24
gemini-1.5-pro-exp-0827
Source run ↗diff-fenced66.9%Compare - Filtered rank 25
gpt-4-0125-preview
Source run ↗udiff66.2%Compare - Filtered rank 26
gpt-4-0314
Source run ↗diff66.2%Compare - Filtered rank 27
llama-3.1-405b-instruct (whole)
Source run ↗whole66.2%Compare - Filtered rank 28
Mistral Large (2411)
Source run ↗diff65.4%Compare - Filtered rank 29
gemini-1.5-pro-002
Source run ↗diff-fenced65.4%Compare - Filtered rank 30
gpt-4-1106-preview
Source run ↗udiff65.4%Compare - Filtered rank 31
openrouter/qwen/qwen-2.5-coder-32b-instruct
Source run ↗diff65.4%Compare - Filtered rank 32
qwen-2.5-72b-instruct (bf16)
Source run ↗diff65.4%Compare - Filtered rank 33
yi-lightning
Source run ↗whole65.4%Compare - Filtered rank 34
gpt-4-turbo-2024-04-09 (udiff)
Source run ↗udiff63.9%Compare - Filtered rank 35
llama-3.1-405b-instruct (diff)
Source run ↗diff63.9%Compare - Filtered rank 36
nousresearch/hermes-3-llama-3.1-405b
Source run ↗whole63.9%Compare - Filtered rank 37
ollama/Qwen2.5.1-Coder-7B-Instruct-GGUF:Q8_0-32k
Source run ↗whole63.9%Compare - Filtered rank 38
ollama/qwen2.5-coder:14b
Source run ↗whole61.7%Compare - Filtered rank 39
gemini-exp-1114
Source run ↗diff60.9%Compare - Filtered rank 40
Mistral Large 2 (2407)
Source run ↗whole60.2%Compare - Filtered rank 41
llama-3.3-70b-instruct
Source run ↗diff59.4%Compare - Filtered rank 42
Grok-2
Source run ↗whole58.6%Compare - Filtered rank 43
llama-3.1-70b-instruct
Source run ↗whole58.6%Compare - Filtered rank 44
ollama/qwen2.5:32b-instruct-q8_0
Source run ↗whole58.6%Compare - Filtered rank 45
Qwen2.5-Coder-7B-Instruct
Source run ↗whole57.9%Compare - Filtered rank 46
gemini-exp-1121
Source run ↗diff57.9%Compare - Filtered rank 47
gpt-3.5-turbo-0301
Source run ↗whole57.9%Compare - Filtered rank 48
gemini-1.5-pro-001
Source run ↗diff-fenced57.1%Compare - Filtered rank 49
Qwen2 72B Instruct
Source run ↗whole55.6%Compare - Filtered rank 50
gpt-4o-mini
Source run ↗whole55.6%Compare - Filtered rank 51
Grok-2-mini
Source run ↗whole54.9%Compare - Filtered rank 52
Llama-3.1-Nemotron-70B-Instruct-HF
Source run ↗whole54.9%Compare - Filtered rank 53
claude-3-sonnet-20240229
Source run ↗whole54.9%Compare - Filtered rank 54
Nova Pro
Source run ↗whole54.1%Compare - Filtered rank 55
Yi Coder 9B Chat
Source run ↗whole54.1%Compare - Filtered rank 56
ollama/qwen2.5:32b
Source run ↗whole54.1%Compare - Filtered rank 57
gemini-1.5-flash-exp-0827
Source run ↗whole52.6%Compare - Filtered rank 58
qwen2.5-coder:7b-instruct-q8_0
Source run ↗whole51.9%Compare - Filtered rank 59
codestral-2405
Source run ↗whole51.1%Compare - Filtered rank 60
gemini-1.5-flash-002 (0924)
Source run ↗whole51.1%Compare - Filtered rank 61
gpt-3.5-turbo-0125
Source run ↗whole50.4%Compare - Filtered rank 62
gpt-3.5-turbo-0613
Source run ↗whole50.4%Compare - Filtered rank 63
qwen2:72b-instruct-q8_0
Source run ↗whole49.6%Compare - Filtered rank 64
Codestral-22B-v0.1-Q4_K_M
Source run ↗whole48.1%Compare - Filtered rank 65
codestral:22b-v0.1-q8_0
Source run ↗whole48.1%Compare - Filtered rank 66
claude-3-haiku-20240307
Source run ↗whole47.4%Compare - Filtered rank 67
ollama/codestral
Source run ↗whole45.9%Compare - Filtered rank 68
yi-coder:9b-chat-q4_0
Source run ↗whole45.1%Compare - Filtered rank 69
WizardLM-2 8x22B
Source run ↗whole44.4%Compare - Filtered rank 70
gemini-1.5-flash-latest
Source run ↗whole44.4%Compare - Filtered rank 71
ollama/yi-coder:9b-chat-fp16
Source run ↗whole43.6%Compare - Filtered rank 72
Reflection-70B
Source run ↗whole42.1%Compare - Filtered rank 73
Qwen2.5-Coder-3B-Instruct
Source run ↗whole39.1%Compare - Filtered rank 74
Command R (08-24)
Source run ↗whole38.3%Compare - Filtered rank 75
Command R+ (08-24)
Source run ↗whole38.3%Compare - Filtered rank 76
gemini-1.5-flash-8b-exp-0827
Source run ↗whole38.3%Compare - Filtered rank 77
gemini-1.5-flash-8b-exp-0924
Source run ↗whole38.3%Compare - Filtered rank 78
ollama/mistral-small
Source run ↗whole38.3%Compare - Filtered rank 79
llama-3.1-8b-instruct
Source run ↗whole37.6%Compare - Filtered rank 80
qwen1.5-110b-chat
Source run ↗whole37.6%Compare - Filtered rank 81
gemma2:27b-instruct-q8_0
Source run ↗whole36.1%Compare - Filtered rank 82
codeqwen:7b-chat-v1.5-q8_0
Source run ↗whole34.6%Compare - Filtered rank 83
ollama/mistral-nemo:12b-instruct-2407-q4_K_M
Source run ↗whole33.1%Compare - Filtered rank 84
ollama/codegeex4
Source run ↗whole32.3%Compare - Filtered rank 85
Qwen2.5-Coder-1.5B-Instruct
Source run ↗whole31.6%Compare - Filtered rank 86
command-r-plus
Source run ↗whole31.6%Compare - Filtered rank 87
ollama/hermes3:8b-llama3.1-fp16
Source run ↗whole30.1%Compare - Filtered rank 88
ollama/wojtek/opencodeinterpreter:6.7b
Source run ↗whole30.1%Compare - Filtered rank 89
ollama/llama3.2:3b-instruct-fp16
Source run ↗whole26.3%Compare - Filtered rank 90
ollama/tulu3
Source run ↗whole26.3%Compare - Filtered rank 91
ollama/hermes3
Source run ↗whole22.6%Compare - Filtered rank 92
ollama/granite3-dense:8b
Source run ↗whole20.3%Compare - Filtered rank 93
Qwen2.5-Coder-0.5B-Instruct
Source run ↗whole14.3%Compare
132 tasks
- Filtered rank 1
gpt-3.5-turbo-1106
Source run ↗whole56.1%Compare - Filtered rank 2
llama3-70b-8192
Source run ↗diff49.2%Compare
225 tasks
- Filtered rank 1
o1-mini-2024-09-12
Source run ↗whole27.1%Compare
36 tasks
- Filtered rank 1
o1-mini
Source run ↗diff61.1%Compare
33 tasks
- Filtered rank 1
gpt-4-turbo-2024-04-09 (diff)
Source run ↗diff57.6%Compare