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 • Updated • 1.51k • 35Note currently #1 with error rate of 0.353
deepseek-ai/deepseek-coder-1.3b-base
Text Generation • Updated • 20.6k • 60Note # previous #1 error rate 0.38
stabilityai/stable-code-3b
Text Generation • Updated • 5.33k • 629bigcode/starcoder2-7b
Text Generation • Updated • 25.1k • 153bigcode/starcoder2-3b
Text Generation • Updated • 96.7k • 135
Vipitis/santacoder-finetuned-Shadertoys-fine
Text Generation • Updated • 12 • 1Note has noteable difference between fp16 and fp32, will need to run bf16 likely contaminated
google/gemma-7b
Text Generation • Updated • 269k • 3.03kgoogle/codegemma-2b
Text Generation • Updated • 3.79k • 68
Vipitis/santacoder-finetuned-Shadertoys
Text Generation • Updated • 14 • 2Note likely contaminated
Deci/DeciCoder-1b
Text Generation • Updated • 7.4k • 247Note 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 • Updated • 440k • 885
Salesforce/codegen2-1B_P
Text Generation • Updated • 1.26k • 38Note needs rerun with incomplete_generation tag
Vipitis/santacoder-finetuned-the-stack-glsl
Text Generation • Updated • 8 • 3microsoft/phi-1_5
Text Generation • Updated • 93.2k • 1.31kmicrosoft/phi-1
Text Generation • Updated • 10.3k • 200
microsoft/phi-2
Text Generation • Updated • 335k • 3.23kNote performs the worst with error rate of 0.79
Running4🚀ShaderMatch
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
THUDM/codegeex2-6b
Updated • 291 • 252deepseek-ai/deepseek-coder-5.7bmqa-base
Text Generation • Updated • 789 • 7deepseek-ai/deepseek-coder-6.7b-base
Text Generation • Updated • 11.4k • 80bigcode/gpt_bigcode-santacoder
Text Generation • Updated • 16.2k • 25bigcode/starcoderbase
Text Generation • Updated • 1.82k • 390google/codegemma-7b
Text Generation • Updated • 16.7k • 151aiXcoder/aixcoder-7b-base
Text Generation • Updated • 43 • 48Qwen/CodeQwen1.5-7B
Text Generation • Updated • 11.6k • 90ibm-granite/granite-3b-code-base-2k
Text Generation • Updated • 5.53k • 32mistralai/Codestral-22B-v0.1
Text Generation • Updated • 27k • 1.11kdeepseek-ai/DeepSeek-Coder-V2-Lite-Base
Text Generation • Updated • 26.5k • 51Salesforce/codet5p-2b
Text2Text Generation • Updated • 881 • 33facebook/llm-compiler-7b
Text Generation • Updated • 640 • 119meta-llama/Meta-Llama-3.1-8B
Text Generation • Updated • 568k • 843meta-llama/CodeLlama-7b-hf
Text Generation • Updated • 13k • 5901-ai/Yi-Coder-9B
Text Generation • Updated • 1.19k • 35Qwen/Qwen2.5-Coder-1.5B
Text Generation • Updated • 43 • 6Qwen/Qwen2.5-Coder-7B
Text Generation • Updated • 202 • 5