Instructions to use Jack04810/RAVEC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Jack04810/RAVEC with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Jack04810/RAVEC")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Jack04810/RAVEC", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Jack04810/RAVEC with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Jack04810/RAVEC" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Jack04810/RAVEC", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Jack04810/RAVEC
- SGLang
How to use Jack04810/RAVEC 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 "Jack04810/RAVEC" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Jack04810/RAVEC", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "Jack04810/RAVEC" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Jack04810/RAVEC", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Jack04810/RAVEC with Docker Model Runner:
docker model run hf.co/Jack04810/RAVEC
File size: 3,891 Bytes
377660b bc9139e 377660b bc9139e 377660b bc9139e 377660b 254d03a bc9139e 377660b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 | {
"table": "Table 2",
"metric": "Avg@8 (%)",
"average": "Equal weight over ten benchmarks",
"checkpoint_selection": {
"F0": {
"checkpoint_step": 175,
"weights": "Actor"
},
"M1": {
"checkpoint_step": 100,
"weights": "Actor",
"reason": "Single checkpoint with all fourteen benchmarks; also the highest recorded CLEAN2K score."
}
},
"rows": {
"RAVEC-3B": {
"source_scores_percent": {
"Geometry3K": "33.28",
"MathVista": "59.78",
"We-Math": "61.68",
"MathVerse": "57.43",
"LogicVista": "41.67",
"MMMU-Pro": "28.38",
"MathVerse-V": "54.44",
"C-VQA-Real": "66.5414281170483500",
"MARS-Bench": "53.0454990215264200",
"POPE": "87.9833333333333400"
},
"benchmark_count": 10,
"missing_benchmarks": [],
"mean_percent": "54.4230260471908110",
"display_mean_percent": "54.42"
},
"RAVEC-7B": {
"source_scores_percent": {
"Geometry3K": "43.20",
"MathVista": "66.44",
"We-Math": "70.73",
"MathVerse": "67.53",
"LogicVista": "46.28",
"MMMU-Pro": "36.09",
"MathVerse-V": "65.03",
"C-VQA-Real": "70.69",
"MARS-Bench": "55.99",
"POPE": "87.99"
},
"benchmark_count": 10,
"missing_benchmarks": [],
"mean_percent": "60.997",
"display_mean_percent": "61.00",
"raw_scores_percent": {
"geo3k": 43.19883527454243,
"mathvista": 66.4375,
"wemath": 70.72557471264368,
"mathverse": 67.53440366972477,
"logicvista": 46.28076062639821,
"mmmu-pro": 36.09104046242775,
"mathverse-v": 65.03249235474006,
"c-vqa-real": 70.68821564885496,
"mars-bench": 55.990704500978474,
"pope": 87.99166666666667
},
"raw_mean_percent": 60.997119391697694
}
},
"availability": {
"RAVEC-3B": "included",
"RAVEC-7B": "included; M1 step100 Actor, verified against evaluation-time SHA-256 hashes"
},
"provenance": [
{
"source_in_project": "paper/tables/main_results.tex",
"sha256": "6f7c33a0076b6e649c3bc44a4ee600731d3a8f722be5aacf23f567ded0745361"
},
{
"source_in_project": "paper/scripts/build_experiment_tables.py",
"sha256": "fd2a5c3e3d16d9a2501351729b5d5f6099174bc921f82b15094de83a3627fc95"
},
{
"source_in_project": "paper/data/table_statistics_20261107.json",
"sha256": "5e42aba7c5ebf532153f9a14e45665c589cc611e6a1f9151c64a3ef4a6d2f80d"
},
{
"source_in_project": "paper/data/ravec3b_extra_scores_20261108.json",
"sha256": "0fece581e263c3bbc7fd0054cb7938af45b4088ac7234abb332e05b72bede3a8"
},
{
"source_in_project": "checkpoints/ravec_main/step175/manifest.json",
"sha256": "e95a6a5de5794b5f10f1cbad73f4dd6ed22308dd4a397abf0deda4cce64374e2"
},
{
"source_in_project": "RAVEC_GRPO_7B_RESULTS_20261106.md",
"sha256": "4ecf81e9dfe50c71c16fc51a0c92134807bc41234ad43cec8f471908404b6a35"
},
{
"source_in_project": "runtime/assets/runs/formal_7b_4gpu_20261029_000339/M1/eval_step100_suite14/238fbe483d8179bb/summary.json",
"sha256": "87f95c152d4ebc6d193018d7e1648677d2f375012fd35e00c7e6144fcf797329",
"model_fingerprint": "238fbe483d8179bbbab0ce9f4ec044c09bb43661f4f319698f063beb2f6199d7",
"release_manifest": "RAVEC-7B/checkpoint_manifest.json"
}
],
"precision_note": "3B original seven displayed tasks and 7B tasks use reported precision; 3B CVQA/MARS/POPE use raw scoring precision.",
"weight_label_note": "Historical 3B records use EMA_G3R / EMA labels; current paper and table-generation metadata identify this exact step175 export as Actor/student. File hashes identify the released artifact unambiguously.",
"release_scope": "3B and 7B main-result checkpoints"
}
|