Text Generation
Transformers
Safetensors
English
Portuguese
qwen2
mongodb
code
fim
autocomplete
synthetic-data
conversational
Eval Results (legacy)
text-generation-inference
Instructions to use esilva/KapiCoder-Mongo-Nano with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use esilva/KapiCoder-Mongo-Nano with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="esilva/KapiCoder-Mongo-Nano") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("esilva/KapiCoder-Mongo-Nano") model = AutoModelForCausalLM.from_pretrained("esilva/KapiCoder-Mongo-Nano", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use esilva/KapiCoder-Mongo-Nano with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "esilva/KapiCoder-Mongo-Nano" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "esilva/KapiCoder-Mongo-Nano", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/esilva/KapiCoder-Mongo-Nano
- SGLang
How to use esilva/KapiCoder-Mongo-Nano with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "esilva/KapiCoder-Mongo-Nano" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "esilva/KapiCoder-Mongo-Nano", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "esilva/KapiCoder-Mongo-Nano" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "esilva/KapiCoder-Mongo-Nano", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use esilva/KapiCoder-Mongo-Nano with Docker Model Runner:
docker model run hf.co/esilva/KapiCoder-Mongo-Nano
Download provenance/model_lock.json from esilva/KapiCoder-Mongo-Nano: direct link, hf CLI and curl.
- Browser
- Download file 3.96 kB
-
https://huggingface.co/esilva/KapiCoder-Mongo-Nano/resolve/main/provenance/model_lock.json
- Command line
-
hf download hf://esilva/KapiCoder-Mongo-Nano/provenance/model_lock.json
-
curl -L -o model_lock.json https://huggingface.co/esilva/KapiCoder-Mongo-Nano/resolve/main/provenance/model_lock.json
3.96 kB
| { | |
| "schema": "kapi-hf-model-lock/1", | |
| "model": { | |
| "name": "KapiCoder-Mongo-Nano", | |
| "repo_id": "esilva/KapiCoder-Mongo-Nano", | |
| "generation": "v3", | |
| "release_tag": "v0.1.0-preview", | |
| "run": "sft_v3", | |
| "tier": "T0", | |
| "status": "PREVIEW (private)" | |
| }, | |
| "base_model": { | |
| "repo_id": "Qwen/Qwen2.5-Coder-0.5B", | |
| "revision": "8123ea2e9354afb7ffcc6c8641d1b2f5ecf18301", | |
| "model_sha256": "96217dc3a23c02904459bcfb1dcad72fda4ca5e629182cca700f558803fecbc5", | |
| "license": "Apache-2.0" | |
| }, | |
| "weights": { | |
| "format": "safetensors", | |
| "dtype": "bfloat16", | |
| "combined_sha256": "029aab78dde4ee60c0c1a6cb0e5a055b6a81b79ac8224fe64f117a2e04f27e2c", | |
| "files_sha256": { | |
| "config.json": "1a0023f5853290cf4b94ddb544ef778e8ecf7861f2d3e79e75f90e064863ed9c", | |
| "generation_config.json": "215d4506cf90f3590fe821bee47dc92e99caa06f2c1b36076693b5209980009c", | |
| "merges.txt": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3", | |
| "model.safetensors": "6ff7b2528689b448ad8d06e0451ffae4f70f803bd32dc0ab8552f4a70c230e31", | |
| "tokenizer.json": "c0382117ea329cdf097041132f6d735924b697924d6f6fc3945713e96ce87539", | |
| "tokenizer_config.json": "283ec39d07ebfd1904aa2dee3c7d73524ed96c489fdb3915629900d3ab3ac319", | |
| "vocab.json": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910" | |
| } | |
| }, | |
| "training": { | |
| "method": "full", | |
| "master_dtype": "fp32", | |
| "optimizer": "adamw", | |
| "peak_lr": 2e-05, | |
| "weight_decay": 0.01, | |
| "seed": 20261018, | |
| "lora": null, | |
| "trainable_params": 494032768, | |
| "total_steps": 4704, | |
| "steps_per_epoch": 2352, | |
| "tokens_seen": 129999692, | |
| "elapsed_train_h": 4.381, | |
| "started_utc": "2026-10-25T01:59:41Z", | |
| "finished_utc": "2026-10-25T06:23:50Z", | |
| "git_commit": "ba1b880645d86bb33483c016a34782576e901ba9", | |
| "git_dirty": false, | |
| "train_config_sha256": "8e7ef77567ecdc561c17e6520d952df91bdaa3b33a26964a0979641d09ce68f8", | |
| "config_file_sha256": "54fc8718fb07857ded4cf8937c6b785ef4599545cb001924d8cb67cc41b7ecdc", | |
| "libraries": { | |
| "python": "3.12.10", | |
| "torch": "2.10.0+rocm7.14.0a20260611", | |
| "transformers": "4.57.6", | |
| "peft": "0.17.1", | |
| "tokenizers": "0.22.2", | |
| "safetensors": "0.7.0" | |
| }, | |
| "hardware": "AMD Radeon RX 7800 XT", | |
| "single_training_run": true | |
| }, | |
| "dataset": { | |
| "name": "student corpus v3 (not published)", | |
| "published": false, | |
| "label_source": "mixed: teacher_seq (accepted fim_ctx/completion) | gold (llm, grammar, fim_raw) | replay", | |
| "examples": 238453, | |
| "tokens_per_epoch": 64999846, | |
| "replay_examples": 12618, | |
| "replay_token_share": 0.12, | |
| "corpus_manifest_sha256": "98abe23cb2b562cb96ee4f0730550f452f08843c2ba6c084a216f347365c2d1d", | |
| "ids_sha256": "434f2e2cbe01b80bf9eae45747d91a2660cc152dcbad33f444bef8ea33367cf1", | |
| "rows_sha256": "87bcb71bfa0dc4aa6e8e40ab101dd3d5af7ee1ddfbad629bd9180dbc203e1f46", | |
| "student_corpus_config_sha256": "47f02a0b6a2a1564d520c4137715de95932835d143003eb7204b01201f2189fb", | |
| "accepted_sha256": "2a32311843d3479cfd6ac6210555208bc53d01c19b23887f00a6ab37c1ba97bc" | |
| }, | |
| "evaluation": { | |
| "val_v2c": { | |
| "split": "val", | |
| "n": 2000, | |
| "seed": 20261008, | |
| "sample_ids_sha256": "28ba86be496839a321de0035d6a7e21827a5e8de4362151d80958ee8c798562f", | |
| "dataset_sha256": "af3331481966f3ebfb05d36109b43014cd850f669f6f5d2a524f8da6e9609ec7", | |
| "manifest_sha256": "2fef4f342d8a21eaf6d7e9eb1f223b3c8094d56c8dcd00e64f04bc20b20cd3cb" | |
| }, | |
| "homologation_v1": { | |
| "set_sha256": "aca0aae51533a95d830c1f296d621f9203a7d76475658fb1e2dcc8f389495332", | |
| "n": 6030, | |
| "items_published": false | |
| }, | |
| "contract": { | |
| "name": "editor-context-v1", | |
| "golden_sha256": "06ef5347b128d4dcd11e838ebaa45b30797c5ba30e67222276477e9921ae51c3" | |
| } | |
| }, | |
| "publish_tooling": { | |
| "scripts_publish_sha256": "c31297cc42c0b967e55d196d2e88b48a073d6dc988cd1d097313faafc6b0d45b" | |
| } | |
| } | |