covaga

eplan-2027-lora-v4

README

License: apache-2.0

🚀 Quickstart

Requires transformers>=5 and a Hugging Face token (this repo is private — run hf auth login first). The adapter is a standard PEFT checkpoint, fully Unsloth-compatible.

python

# pip install "transformers>=5" unsloth peftfrom unsloth import FastLanguageModelfrom peft import PeftModel model, tokenizer = FastLanguageModel.from_pretrained(    "Qwen/Qwen3.5-9B",    max_seq_length=4096,    load_in_16bit=True,          # adapter trained in BF16)model = PeftModel.from_pretrained(model, "covaga/eplan-2027-lora-v4")FastLanguageModel.for_inference(model) messages = [{"role": "user", "content":    "Which EPLAN 2027 API class creates a new project? Show a minimal C# script."}] inputs = tokenizer.apply_chat_template(    messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda") out = model.generate(inputs, max_new_tokens=512, temperature=0.2)print(tokenizer.decode(out[0][inputs.shape[1]:], skip_special_tokens=True))

🧠 What it does

EPLAN 2027 API reference Q&A — explains classes, members and namespaces of the EPLAN Electric P8 2027 API (Eplan.EplApi.DataModel, Eplan.EplApi.HEServices, …) in English or Spanish.
C# script generation — produces scripts for EPLAN's scripting engine. Training data constrained output to C# 5-era syntax for safety; EPLAN 2027 actually compiles C# 13 via Roslyn, so you can modernize the generated code freely.
Error diagnosis — explains EPLAN script compile/runtime errors and suggests fixes.
Calibrated refusals — trained to say "this API does not exist" instead of hallucinating plausible-looking identifiers (see benchmarks — still imperfect).
Bilingual EN/ES — questions and answers work in both languages.

📊 Benchmarks

Curated 44-question eval (covaga/eplan-2027-eval): API reference, C# codegen, error diagnosis, unanswerable questions, invented-API traps, EN/ES, multi-constraint prompts. Valid % = cited identifiers that exist in the official EPLAN 2027 docs; invented = identifiers not in the docs; refusals = correct refusals on the 9 trap questions.

Table with columns: Model, IDs cited, Valid IDs, Invented IDs, expect_ids, Forbid viol., Refusals (/9)
Model	IDs cited	Valid IDs	Invented IDs	`expect_ids`	Forbid viol.	Refusals (/9)
Qwen3.5-9B (base)	401	43%	230	33/44	5	1/9
v1 (deleted)	133	59%	55	35/44

¹ This model reaches 5/9 semantically — the scorer's regex misses one genuine refusal.

Clean test split (60 questions, documents never seen in training):

Table with columns: Model, Valid IDs, expect_ids, Forbid viol.
Model	Valid IDs	`expect_ids`	Forbid viol.
v3	63%	60/60	0
v4 — this repo	pending	pending	pending

🧭 Which version should you use?

Table with columns: Version, Character, Recommendation
Version	Character	Recommendation
v4 (this repo)	Most conservative: fewest IDs cited (111) and fewest invented (42)	Pick when fabrication avoidance matters more than coverage
v3	Balanced: solid precision + best refusal behaviour	Default choice for most use cases
v1 / v2	Superseded, deleted	—

💻 Running locally — minimum specs

Table with columns: Build, File size, Minimum, Comfortable, Notes
Build	File size	Minimum	Comfortable	Notes
GGUF Q4_K_M	5.6 GB + 0.9 GB mmproj	~8 GB free RAM	12–16 GB RAM, or GPU with ≥8 GB VRAM	Best option — CPU-friendly via llama.cpp; works on Apple Silicon (`llama-mtmd-cli`)
Merged BF16	18.8 GB	GPU ≥24 GB VRAM	A100/RTX 3090+	CPU inference possible at ~24 GB+ RAM but very slow
LoRA adapter	233 MB (+ base)

Recommended local path: GGUF Q4_K_M through llama.cpp / LM Studio / Ollama-compatible runners. For anything serious, pair it with retrieval (below) — grounding the model with real docs is the single biggest quality win.

🔌 Ground it — RAG & tooling ecosystem

This model answers from memory; pairing it with the official EPLAN 2027 documentation dramatically improves accuracy and cuts hallucinations. The ecosystem around it:

covagashi/eplan-rag-mcp — MCP server + tooling for EPLAN automation: official 2027 action registries, live actions and ribbon labels. Gives an agent real, verified EPLAN identifiers instead of guesses.
covagashi/eplan-cloudflare-rags — Cloudflare-hosted RAG over the official EPLAN 2027 docs (live endpoint rag2027.covaga.xyz, /search + /file). Feed retrieved doc pages into the prompt to ground answers.
covagashi/eplan-development-skill — curated EPLAN development reference (actions, data access, remoting, pitfalls, E3D) usable as an agent skill or as grounded context.

In our eval, retrieval alone lifted refusal accuracy on invented-API traps from 1/9 to 4/9 — combining the adapter with RAG is the intended production setup.

Merged model (no PEFT needed, 18.8 GB BF16): covaga/eplan-2027-9b-merged
GGUF for llama.cpp — Q4_K_M (5.6 GB) + mmproj (0.9 GB): covaga/eplan-2027-9b-merged-GGUF
bash
llama-mtmd-cli -m Qwen3.5-9B.Q4_K_M.gguf --mmproj Qwen3.5-9B.BF16-mmproj.gguf
Eval harness & data: covaga/eplan-2027-eval

Sample prompts

text

EN  Which EPLAN 2027 API class lets me enumerate all pages of the open project?EN  Write a C# script for EPLAN that exports every schematic page to PDF.ES  ¿Qué clase de la API de EPLAN 2027 permite crear un proyecto nuevo?ES  Escribe un script en C# que recorra todas las páginas del proyecto abierto.

⚠️ Limitations

Read before production use.

Still fabricates sometimes. On invented-API traps the model refuses only ~4–5/9 times — it can emit plausible-sounding identifiers that do not exist in EPLAN.

Refusals leak inventions. Even correct refusals occasionally suggest made-up alternative APIs.

Valid identifier ≠ correct code. A real class name can still be used with the wrong signature or semantics.

Always compile in real EPLAN before shipping a generated script; the model targets a C# 5-era subset while EPLAN 2027 compiles C# 13 via Roslyn.

Text-only. The vision tower of Qwen3.5-9B was not trained — no image/screenshot understanding of EPLAN UIs.

Synthetic data risk. The dataset was generated and filtered automatically; upstream errors may have been inherited. Verify against the official EPLAN 2027 documentation.

Base model Qwen/Qwen3.5-9B retains its own license terms.

🏋️ Training details

Table with columns: Setting, Value
Setting	Value
Method	LoRA (PEFT / Unsloth), BF16
Base model	`Qwen/Qwen3.5-9B`
Rank / α / dropout	r=32 · α=32 · 0
Trainable params	58.2M
Target modules	Full attention + MLP — Unsloth `finetune_*` filters exclude DeltaNet-specific projections
Sequence length	2,048 tokens, loss on responses only
Data	— 17,328 rows, document-level split — plus generated refusal/ambiguity/error examples verified against the official EPLAN 2027 docs

📚 Dataset

Trained on covaga/eplan-2027-api-qa (public): 17,328 QA/codegen rows built from the official EPLAN Electric P8 2027 API documentation, split at document level so no document leaks between train/test.

📝 Citation

bibtex

@misc{eplan-2027-lora-v4,  author       = {covaga},  title        = {EPLAN Electric P8 2027 API Assistant — Qwen3.5-9B LoRA (v4)},  year         = {2025},  publisher    = {Hugging Face},  howpublished = {\url{https://huggingface.co/covaga/eplan-2027-lora-v4}}}

🇪🇸 Español

Adaptador LoRA (BF16, r=32) sobre Qwen/Qwen3.5-9B que convierte el modelo en un asistente de la API de EPLAN Electric P8 2027: responde preguntas sobre clases y miembros de Eplan.EplApi.*, genera scripts de C# para el motor de scripting de EPLAN, diagnostica errores de compilación/ejecución y rechaza inventar APIs no documentadas. Bilingüe ES/EN. v4 es la variante más conservadora (menos identificadores citados e inventados); para mejor equilibrio precisión/negativas usa eplan-2027-lora-v3. Compila siempre los scripts en EPLAN real antes de producción — puede seguir inventando identificadores plausibles en una minoría de casos.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

covaga

Model Tree

Base

Qwen/Qwen3.5-9B

Adapter

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

🚀 Quickstart

Requires transformers>=5 and a Hugging Face token (this repo is private — run hf auth login first). The adapter is a standard PEFT checkpoint, fully Unsloth-compatible.

python

# pip install "transformers>=5" unsloth peftfrom unsloth import FastLanguageModelfrom peft import PeftModel model, tokenizer = FastLanguageModel.from_pretrained(    "Qwen/Qwen3.5-9B",    max_seq_length=4096,    load_in_16bit=True,          # adapter trained in BF16)model = PeftModel.from_pretrained(model, "covaga/eplan-2027-lora-v4")FastLanguageModel.for_inference(model) messages = [{"role": "user", "content":    "Which EPLAN 2027 API class creates a new project? Show a minimal C# script."}] inputs = tokenizer.apply_chat_template(    messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda") out = model.generate(inputs, max_new_tokens=512, temperature=0.2)print(tokenizer.decode(out[0][inputs.shape[1]:], skip_special_tokens=True))

🧠 What it does

EPLAN 2027 API reference Q&A — explains classes, members and namespaces of the EPLAN Electric P8 2027 API (Eplan.EplApi.DataModel, Eplan.EplApi.HEServices, …) in English or Spanish.
C# script generation — produces scripts for EPLAN's scripting engine. Training data constrained output to C# 5-era syntax for safety; EPLAN 2027 actually compiles C# 13 via Roslyn, so you can modernize the generated code freely.
Error diagnosis — explains EPLAN script compile/runtime errors and suggests fixes.
Calibrated refusals — trained to say "this API does not exist" instead of hallucinating plausible-looking identifiers (see benchmarks — still imperfect).
Bilingual EN/ES — questions and answers work in both languages.

📊 Benchmarks

Table with columns: Model, IDs cited, Valid IDs, Invented IDs, expect_ids, Forbid viol., Refusals (/9)
Model	IDs cited	Valid IDs	Invented IDs	`expect_ids`	Forbid viol.	Refusals (/9)
Qwen3.5-9B (base)	401	43%	230	33/44	5	1/9
v1 (deleted)	133	59%	55	35/44

¹ This model reaches 5/9 semantically — the scorer's regex misses one genuine refusal.

Clean test split (60 questions, documents never seen in training):

Table with columns: Model, Valid IDs, expect_ids, Forbid viol.
Model	Valid IDs	`expect_ids`	Forbid viol.
v3	63%	60/60	0
v4 — this repo	pending	pending	pending

🧭 Which version should you use?

Table with columns: Version, Character, Recommendation
Version	Character	Recommendation
v4 (this repo)	Most conservative: fewest IDs cited (111) and fewest invented (42)	Pick when fabrication avoidance matters more than coverage
v3	Balanced: solid precision + best refusal behaviour	Default choice for most use cases
v1 / v2	Superseded, deleted	—

💻 Running locally — minimum specs

Table with columns: Build, File size, Minimum, Comfortable, Notes
Build	File size	Minimum	Comfortable	Notes
GGUF Q4_K_M	5.6 GB + 0.9 GB mmproj	~8 GB free RAM	12–16 GB RAM, or GPU with ≥8 GB VRAM	Best option — CPU-friendly via llama.cpp; works on Apple Silicon (`llama-mtmd-cli`)
Merged BF16	18.8 GB	GPU ≥24 GB VRAM	A100/RTX 3090+	CPU inference possible at ~24 GB+ RAM but very slow
LoRA adapter	233 MB (+ base)

🔌 Ground it — RAG & tooling ecosystem

This model answers from memory; pairing it with the official EPLAN 2027 documentation dramatically improves accuracy and cuts hallucinations. The ecosystem around it:

covagashi/eplan-rag-mcp — MCP server + tooling for EPLAN automation: official 2027 action registries, live actions and ribbon labels. Gives an agent real, verified EPLAN identifiers instead of guesses.
covagashi/eplan-cloudflare-rags — Cloudflare-hosted RAG over the official EPLAN 2027 docs (live endpoint rag2027.covaga.xyz, /search + /file). Feed retrieved doc pages into the prompt to ground answers.
covagashi/eplan-development-skill — curated EPLAN development reference (actions, data access, remoting, pitfalls, E3D) usable as an agent skill or as grounded context.

In our eval, retrieval alone lifted refusal accuracy on invented-API traps from 1/9 to 4/9 — combining the adapter with RAG is the intended production setup.

Merged model (no PEFT needed, 18.8 GB BF16): covaga/eplan-2027-9b-merged
GGUF for llama.cpp — Q4_K_M (5.6 GB) + mmproj (0.9 GB): covaga/eplan-2027-9b-merged-GGUF
bash
llama-mtmd-cli -m Qwen3.5-9B.Q4_K_M.gguf --mmproj Qwen3.5-9B.BF16-mmproj.gguf
Eval harness & data: covaga/eplan-2027-eval

Sample prompts

text

EN  Which EPLAN 2027 API class lets me enumerate all pages of the open project?EN  Write a C# script for EPLAN that exports every schematic page to PDF.ES  ¿Qué clase de la API de EPLAN 2027 permite crear un proyecto nuevo?ES  Escribe un script en C# que recorra todas las páginas del proyecto abierto.

⚠️ Limitations

Read before production use.

Still fabricates sometimes. On invented-API traps the model refuses only ~4–5/9 times — it can emit plausible-sounding identifiers that do not exist in EPLAN.

Refusals leak inventions. Even correct refusals occasionally suggest made-up alternative APIs.

Valid identifier ≠ correct code. A real class name can still be used with the wrong signature or semantics.

Always compile in real EPLAN before shipping a generated script; the model targets a C# 5-era subset while EPLAN 2027 compiles C# 13 via Roslyn.

Text-only. The vision tower of Qwen3.5-9B was not trained — no image/screenshot understanding of EPLAN UIs.

Synthetic data risk. The dataset was generated and filtered automatically; upstream errors may have been inherited. Verify against the official EPLAN 2027 documentation.

Base model Qwen/Qwen3.5-9B retains its own license terms.

🏋️ Training details

Table with columns: Setting, Value
Setting	Value
Method	LoRA (PEFT / Unsloth), BF16
Base model	`Qwen/Qwen3.5-9B`
Rank / α / dropout	r=32 · α=32 · 0
Trainable params	58.2M
Target modules	Full attention + MLP — Unsloth `finetune_*` filters exclude DeltaNet-specific projections
Sequence length	2,048 tokens, loss on responses only
Data	— 17,328 rows, document-level split — plus generated refusal/ambiguity/error examples verified against the official EPLAN 2027 docs

📚 Dataset

📝 Citation

bibtex

@misc{eplan-2027-lora-v4,  author       = {covaga},  title        = {EPLAN Electric P8 2027 API Assistant — Qwen3.5-9B LoRA (v4)},  year         = {2025},  publisher    = {Hugging Face},  howpublished = {\url{https://huggingface.co/covaga/eplan-2027-lora-v4}}}

eplan-2027-lora-v4

README

🚀 Quickstart

🧠 What it does

📊 Benchmarks

🧭 Which version should you use?

💻 Running locally — minimum specs

🔌 Ground it — RAG & tooling ecosystem

📦 Usage notes & related artifacts

⚠️ Limitations

🏋️ Training details

📚 Dataset

📝 Citation

🇪🇸 Español

Explore FriendliAI today

README

🚀 Quickstart

🧠 What it does

📊 Benchmarks

🧭 Which version should you use?

💻 Running locally — minimum specs

🔌 Ground it — RAG & tooling ecosystem

📦 Usage notes & related artifacts

⚠️ Limitations

🏋️ Training details

📚 Dataset

📝 Citation

🇪🇸 Español