Model reference · open weights
bloomz is an open-weight language model from bigscience, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
Table of Contents 1. Model Summary 2. Use 3. Limitations 4. Training 5. Evaluation 7. Citation Model Summary We present BLOOMZ & mT0, a family of models capable of following human instructions in dozens of languages zero-shot. We finetune BLOOM & mT5 pretrained multilingual language models on our crosslingual task mixture (xP3) and find the resulting models capable of crosslingual generalization to unseen tasks & languages. - Repository: bigscience-workshop/xmtf - Paper: Crosslingual Generalization through Multitask Finetuning - Point of Contact: Niklas Muennighoff - Languages: Refer to bloom for pretraining & xP3 for finetuning language proportions. It understands both pretraining & finetuning languages. - BLOOMZ & mT0 Model Family: Use Intended use We recommend using the model to perform tasks expressed in natural language. For example, given the prompt "Translate to English: Je t’aime.", the model will most likely answer "I love you.". Some prompt ideas from our paper: - 一个传奇的开端，一个不灭的神话，这不仅仅是一部电影，而是作为一个走进新时代的标签，永远彪炳史册。你认为这句话的立场是赞扬、中立还是批评? - Suggest at least five related search terms to "Mạng neural nhân tạo". - Write a fairy tale about a troll saving a princess from a dangerous dragon. The fairy tale is a masterpiece that has achieved praise worldwide and its moral is "Heroes Come in All Shapes and Sizes". Story (in Spanish): - Explain in a sentence in Telugu what is backpropagation in neural networks. Feel free to share your generations in the Community tab! How to use CPU GPU GPU in 8bit Limitations Prompt Engineering: The performance may vary depending on the prompt. For BLOOMZ models, we recommend making it very clear when the input stops to avoid the model trying to continue it. For example, the prompt "Translate to English: Je t'aime" without the full stop (.) at the end, may result in the model trying to continue the French sentence. Better prompts are e.g. "Translate to English: Je t'aime.", "Translate to English: Je t'aime. Translation:" "What is "Je t'aime." in English?", where it is clear for the model when it should answer. Further, we recommend providing the model as much context as possible. For example, if you want it to answer in Telugu, then
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | bigscience |
|---|---|
| Type | Language models |
| Parameters (lead) | 559M |
| Variants | 2 |
| Runs with | transformers |
| Released | 2022-10-08 |
| Popularity | 961k downloads / month |
| Likes | 138 |
| Licence | Commercial licence needed |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Coreference resolution | Winogrande XL (xl) | Accuracy | 52.41 |
| Coreference resolution | XWinograd (en) | Accuracy | 51.01 |
| Coreference resolution | XWinograd (fr) | Accuracy | 51.81 |
| Coreference resolution | XWinograd (jp) | Accuracy | 52.03 |
| Coreference resolution | XWinograd (pt) | Accuracy | 53.99 |
| Coreference resolution | XWinograd (ru) | Accuracy | 53.97 |
| Coreference resolution | XWinograd (zh) | Accuracy | 54.76 |
| Natural language inference | ANLI (r1) | Accuracy | 33.4 |
| Natural language inference | ANLI (r2) | Accuracy | 33.4 |
| Natural language inference | ANLI (r3) | Accuracy | 33.5 |
| Natural language inference | SuperGLUE (cb) | Accuracy | 53.57 |
| Natural language inference | SuperGLUE (rte) | Accuracy | 67.15 |
| Natural language inference | XNLI (ar) | Accuracy | 44.46 |
| Natural language inference | XNLI (bg) | Accuracy | 39.76 |
| Natural language inference | XNLI (de) | Accuracy | 39.36 |
| Natural language inference | XNLI (el) | Accuracy | 40.96 |
| Natural language inference | XNLI (en) | Accuracy | 46.43 |
| Natural language inference | XNLI (es) | Accuracy | 44.98 |
| Natural language inference | XNLI (fr) | Accuracy | 45.54 |
| Natural language inference | XNLI (hi) | Accuracy | 41.81 |
| Natural language inference | XNLI (ru) | Accuracy | 39.64 |
| Natural language inference | XNLI (sw) | Accuracy | 38.35 |
| Natural language inference | XNLI (th) | Accuracy | 35.5 |
| Natural language inference | XNLI (tr) | Accuracy | 37.31 |
Using it via the API
Once AxForge deploys bloomz for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (bloomz below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/chat/completions \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"bloomz","messages":[{"role":"user","content":"Hello"}]}'
Licence
The weights are open but bigscience-bloom-rail-1.0 needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗