GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Pablo-Flores-Mollinedo
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Pablo-Flores-Mollinedo
Model Tree
Qwen/Qwen2.5-Coder-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
This adapter uses VerilogEval-derived repair data. VerilogEval v2 numbers are therefore benchmark-targeted repair results, not clean zero-shot leaderboard placement.
For clean/public production use, v4.1 remains the recommended adapter unless you specifically want VerilogEval-targeted repair behavior.
text
original spec+ incorrect generated code+ compile/simulation failure hint-> corrected verified TopModuleSpec-to-RTL, 156 tasks, n=1, temperature=0, top_p=0.01.
| Adapter | Compile | Functional |
|---|---|---|
| v4.1 | 82.05% | 36.54% |
| v5b | 85.26% | 36.54% |
| v6 | 84.62% | 37.82% |
v6 vs v4.1:
text
compile: +2.56 pointsfunctional: +1.28 pointstext
data/v6_error_driven_repair.jsonl508 examples96 validated repairspython
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModel base = "Qwen/Qwen2.5-Coder-7B-Instruct"adapter = "Pablo-Flores-Mollinedo/verilog-qwen2.5-coder-7b-v6-error-driven-repair-lora" tokenizer = AutoTokenizer.from_pretrained(base)model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")model = PeftModel.from_pretrained(model, adapter)from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModel base = "Qwen/Qwen2.5-Coder-7B-Instruct"adapter = "Pablo-Flores-Mollinedo/verilog-qwen2.5-coder-7b-v6-error-driven-repair-lora" tokenizer = AutoTokenizer.from_pretrained(base)model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")model = PeftModel.from_pretrained(model, adapter)