models to evaluate
collecting models I want to evaluate on shadereval-task2: https://github.com/bigcode-project/bigcode-evaluation-harness/pull/173 at fp16!!
Text Generation • 7B • Updated • 705 • 50Note currently #1 with error rate of 0.353
deepseek-ai/deepseek-coder-1.3b-base
Text Generation • Updated • 16.8k • 110Note # previous #1 error rate 0.38
-
stabilityai/stable-code-3b
Text Generation • 3B • Updated • 10.6k • 659 -
bigcode/starcoder2-7b
Text Generation • 7B • Updated • 10.1k • 215 -
bigcode/starcoder2-3b
Text Generation • 3B • Updated • 114k • 220
Vipitis/santacoder-finetuned-Shadertoys-fine
Text Generation • 1B • Updated • 11Note has noteable difference between fp16 and fp32, will need to run bf16 likely contaminated
-
google/gemma-7b
Text Generation • 9B • Updated • 105k • • 3.33k -
google/codegemma-2b
Text Generation • 3B • Updated • 32.8k • • 99
Vipitis/santacoder-finetuned-Shadertoys
Text Generation • 1B • Updated • 20 • 2Note likely contaminated
Deci/DeciCoder-1b
Text Generation • 1B • Updated • 535 • 248Note current result is not fully correct, need to rerun model - however I don't know a transformers version that runs without errors
-
google/gemma-2b
Text Generation • 3B • Updated • 332k • • 1.18k
Salesforce/codegen2-1B_P
Text Generation • Updated • 1.06k • 41Note needs rerun with incomplete_generation tag
-
Vipitis/santacoder-finetuned-the-stack-glsl
Text Generation • 1B • Updated • 28 • 2 -
microsoft/phi-1_5
Text Generation • 1B • Updated • 64.5k • • 1.36k -
microsoft/phi-1
Text Generation • 1B • Updated • 9.12k • • 220
microsoft/phi-2
Text Generation • 3B • Updated • 536k • • 3.46kNote performs the worst with error rate of 0.79
ShaderMatch
🚀11code completion benchmark for GLSL shadercode
Note this space holds the evaluation metric that is used. It also has a usually up to date leaderboard. check for updates: https://huggingface.co/spaces/Vipitis/shadermatch/blob/main/result_preview.png
-
zai-org/codegeex2-6b
Updated • 125 • 258 -
deepseek-ai/deepseek-coder-5.7bmqa-base
Text Generation • Updated • 1.3k • 10 -
deepseek-ai/deepseek-coder-6.7b-base
Text Generation • 7B • Updated • 48.3k • 123 -
bigcode/gpt_bigcode-santacoder
Text Generation • 1B • Updated • 39.9k • 27 -
bigcode/starcoderbase
Text Generation • Updated • 12.3k • 418 -
google/codegemma-7b
Text Generation • Updated • 2.9k • 219 -
aiXcoder/aixcoder-7b-base
Text Generation • 7B • Updated • 808 • 55 -
Qwen/CodeQwen1.5-7B
Text Generation • 7B • Updated • 492 • 103 -
ibm-granite/granite-3b-code-base-2k
Text Generation • 3B • Updated • 35k • 38 -
mistralai/Codestral-22B-v0.1
22B • Updated • 12.5k • 1.33k -
deepseek-ai/DeepSeek-Coder-V2-Lite-Base
Text Generation • 16B • Updated • 3.28k • 111 -
Salesforce/codet5p-2b
Updated • 188 • 35 -
facebook/llm-compiler-7b
Text Generation • Updated • 41 • 140 -
meta-llama/Llama-3.1-8B
Text Generation • 8B • Updated • 1.24M • • 2.21k -
meta-llama/CodeLlama-7b-hf
Text Generation • Updated • 3.91k • 122 -
01-ai/Yi-Coder-9B
Text Generation • 9B • Updated • 8.65k • 45 -
Qwen/Qwen2.5-Coder-1.5B
Text Generation • 2B • Updated • 311k • • 92 -
Qwen/Qwen2.5-Coder-7B
Text Generation • 8B • Updated • 566k • • 146 -
infly/OpenCoder-1.5B-Base
Text Generation • 2B • Updated • 1.92k • 24 -
Qwen/Qwen2.5-Coder-0.5B
Text Generation • 0.5B • Updated • 31.5k • • 54