Model reference · open weights
HARC-Qwen2.5 is an open-weight language model from microsoft, 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
HARC — Qwen2.5-7B-Instruct HARC safety-alignment LoRA merged into Qwen/Qwen2.5-7B-Instruct (full standalone model). Part of the HARC release; see paper arXiv:2607.00572.
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | microsoft |
|---|---|
| Type | Language models |
| Parameters (lead) | 7.6B |
| Variants | 1 |
| Runs with | transformers |
| Based on | Qwen/Qwen2.5-7B-Instruct |
| Released | 2026-07-02 |
| Popularity | 517 downloads / month |
| Likes | 1 |
| Licence | Open weights |
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.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| HARC-Qwen2.5-7B-Instruct | 7.6B | BF16 | ~17.5 GB | ✓ | Weights ↗ |
Using it via the API
Once AxForge deploys harc-qwen2-5 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (harc-qwen2-5 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":"harc-qwen2-5","messages":[{"role":"user","content":"Hello"}]}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗