Model reference · open weights
Llama-3.1-Tulu-3-SFT is an open-weight language model from allenai. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Maker | allenai |
|---|---|
| Type | Language models |
| Task | Text gen |
| Parameters (lead) | 8.0B |
| Context | 128k tokens |
| Runs with | transformers |
| Based on | meta-llama/Llama-3.1-8B |
| Released | 2024-11-18 |
| Popularity | 19k downloads / month |
| Licence | Open, with conditions |
About
Tülu3 is a leading instruction following model family, offering fully open-source data, code, and recipes designed to serve as a comprehensive guide for modern post-training techniques. Tülu3 is designed for state-of-the-art performance on a diversity of tasks in addition to chat, such as MATH, GSM8K, and IFEval.
| Stage | Llama 3.1 8B | Llama 3.1 70B |
|---|---|---|
| Base Model | meta-llama/Llama-3.1-8B | meta-llama/Llama-3.1-70B |
| SFT | allenai/Llama-3.1-Tulu-3-8B-SFT | allenai/Llama-3.1-Tulu-3-70B-SFT |
| DPO | allenai/Llama-3.1-Tulu-3-8B-DPO | allenai/Llama-3.1-Tulu-3-70B-DPO |
| Final Models (RLVR) | allenai/Llama-3.1-Tulu-3-8B | allenai/Llama-3.1-Tulu-3-70B |
| Reward Model (RM) | allenai/Llama-3.1-Tulu-3-8B-RM | (Same as 8B) |
| Stage | Llama 3.1 405B |
|---|---|
| Base Model | meta-llama/llama-3.1-405B |
| SFT | allenai/llama-3.1-Tulu-3-405B-SFT |
| DPO | allenai/llama-3.1-Tulu-3-405B-DPO |
| Final Model (RLVR) | allenai/llama-3.1-Tulu-3-405B |
| Reward Model (RM) | (Same as 8B) |
To load the model with HuggingFace, use the following snippet:
from transformers import AutoModelForCausalLM
tulu_model = AutoModelForCausalLM.from_pretrained("allenai/Llama-3.1-Tulu-3-8B-SFT")
As a Llama base model, the model can be easily served with:
vllm serve allenai/Llama-3.1-Tulu-3-8B-SFT
Note that given the long chat template of Llama, you may want to use --max_model_len=8192.
The chat template for our models is formatted as:
Or with new lines expanded:
How are you doing?
I'm just a computer program, so I don't have feelings, but I'm functioning as expected. How can I assist you today?
It is embedded within the tokenizer as well, for tokenizer.apply_chat_template.
In Ai2 demos, we use this system prompt by default:
You are Tulu 3, a helpful and harmless AI Assistant built by the Allen Institute for AI.
The model has not been trained with a specific system prompt in mind.
The Tülu3 models have limited safety training, but are not deployed automatically with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so). It is also unknown what the size and composition of the corpus was used to train the base Llama 3.1 models, however it is likely to have included a mix of Web data and technical sources like books and code. See the Falcon 180B model card for an example of this.
| Benchmark (eval) | Tülu 3 SFT 8B | Tülu 3 DPO 8B | Tülu 3 8B | Llama 3.1 8B Instruct | Qwen 2.5 7B Instruct | Magpie 8B | Gemma 2 9B Instruct | Ministral 8B Instruct |
|---|---|---|---|---|---|---|---|---|
| Avg. | 60.4 | 64.4 | 64.8 | 62.2 | 57.8 | 44.7 | 55.2 | 58.3 |
| MMLU (0 shot, CoT) | 65.9 | 68.7 | 68.2 | 71.2 | 76.6 | 62.0 | 74.6 | 68.5 |
| PopQA (15 shot) | 29.3 | 29.3 | 29.1 | 20.2 | 18.1 | 22.5 | 28.3 | 20.2 |
| TruthfulQA (6 shot) | 46.8 | 56.1 | 55.0 | 55.1 | 63.1 | 57.0 | 61.4 | 55.5 |
| BigBenchHard (3 shot, CoT) | 67.9 | 65.8 | 66.0 | 62.8 | 21.7 | 0.9 | 2.5 | 56.2 |
| DROP (3 shot) | 61.3 | 62.5 | **6 |
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys llama-3-1-tulu-3-sft for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (llama-3-1-tulu-3-sft 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":"llama-3-1-tulu-3-sft","messages":[{"role":"user","content":"Hello"}]}'
Create an account — your API key is available in the console. 5M tokens/month currently included with every new account at launch.