Model reference · open weights
airoboros-mistral2.2 is an open-weight language model from teknium. 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 | teknium |
|---|---|
| Type | Language models |
| Task | Text gen |
| Context | 32k tokens |
| Runs with | transformers |
| Based on | mistralai/Mistral-7b-V0.1 |
| Released | 2023-10-03 |
| Popularity | 24 downloads / month |
| Licence | Open weights |
About
Mistral trained with the airoboros dataset!
Actual dataset is airoboros 2.2, but it seems to have been replaced on hf with 2.2.1.
Prompt Format:
USER:
ASSISTANT:
TruthfulQA:
hf-causal-experimental (pretrained=/home/teknium/dakota/lm-evaluation-harness/airoboros2.2-mistral/,dtype=float16), limit: None, provide_description: False, num_fewshot: 0, batch_size: 8
| Task |Version|Metric|Value | |Stderr|
|-------------|------:|------|-----:|---|-----:|
|truthfulqa_mc| 1|mc1 |0.3562|± |0.0168|
| | |mc2 |0.5217|± |0.0156|
Wandb training charts: https://wandb.ai/teknium1/airoboros-mistral-7b/runs/airoboros-mistral-1?workspace=user-teknium1
More info to come
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys airoboros-mistral2-2 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (airoboros-mistral2-2 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":"airoboros-mistral2-2","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.