GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
burkimbia
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
burkimbia
Model Tree
mistralai/Mistral-7B-Instruct-v0.3
Input Modalities
Output Modalities
Supported Functionality
| Repository | Contents | Use it for |
|---|---|---|
BIA-MISTRAL-7B-SACHI (this repo) | LoRA adapter (~100 MB) | applying the fine-tune on top of your own base copy |
BIA-MISTRAL-7B-SACHI_merged | full weights, bf16, 14.5 GB | vLLM, TGI, any fp16/bf16 server |
BIA-MISTRAL-7B-SACHI_4bit | full weights, bitsandbytes 4-bit, 7.2 GB | transformers on a 12 GB GPU |
Serving in production? Take _merged and read its card: it documents the vLLM
setup, the prefix-caching flag and the batching behaviour.
The model was fine-tuned on a raw completion prompt, not on a chat template.
Do not wrap it in apply_chat_template. Use this exact string:
text
<s>You are an expert Moore translator. Translate the provided {SRC} text to {TGT}.The Moore alphabet is: a, ã, b, d, e, ẽ, ɛ, f, g, h, i, ĩ, ɩ, k, l, m, n, o, õ, p, r, s, t, u, ũ, ʋ, v, w, y, z.Based on source language ({SRC}), provide the {TGT} text.[INST]### {SRC}:{TEXT}[/INST]### {TGT}:
{SRC} and {TGT} are the English words French and Moore, in either
direction. The leading blank line and the <s> are part of the training format;
keep them.
python
import torchfrom peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizerBASE = "mistralai/Mistral-7B-Instruct-v0.3"ADAPTER = "burkimbia/BIA-MISTRAL-7B-SACHI"tok = AutoTokenizer.from_pretrained(ADAPTER)tok.padding_side, tok.pad_token = "left", tok.eos_tokenmodel = AutoModelForCausalLM.from_pretrained(BASE, torch_dtype=torch.bfloat16, device_map="auto")model = PeftModel.from_pretrained(model, ADAPTER)PROMPT = """<s>You are an expert Moore translator. Translate the provided {src} text to {tgt}.The Moore alphabet is: a, ã, b, d, e, ẽ, ɛ, f, g, h, i, ĩ, ɩ, k, l, m, n, o, õ, p, r, s, t, u, ũ, ʋ, v, w, y, z.Based on source language ({src}), provide the {tgt} text.[INST]### {src}:{text}[/INST]### {tgt}:"""def translate(text, src="French", tgt="Moore"):inputs = tok(PROMPT.format(src=src, tgt=tgt, text=text), return_tensors="pt").to(model.device)with torch.no_grad():out = model.generate(**inputs, max_new_tokens=256, do_sample=False)return tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True).strip()print(translate("Le marché est fermé aujourd'hui."))print(translate("Raagã pagame rũndã.", src="Moore", tgt="French"))
Decoding: do_sample=False. The model is a translator, not a chat assistant,
and sampling only adds noise.
To merge the adapter into standalone weights yourself:
python
merged = model.merge_and_unload()merged.save_pretrained("bia-mistral-sachi-merged")
_merged and _4bit are exactly that, already done and published.
burkimbia/BIA-NLLB-600M-11E — much smaller seq2seq translator, better latency, lower quality on long sentencesburkimbia/BIA-WHISPER-LARGE-SACHI_V2 — Mooré speech recognitionburkimbia/BIA-SPARKTTS-V4 — Mooré text to speechBenchmark and leaderboard: burkimbia/mt-benchmark-public,
scored on BLEU, METEOR and chrF across five themes (administrative, daily life,
health, religious, arts and technology).
### Moore: marker changes the output quality.bibtex
@misc{bia-mistral-sachi,title = {BIA-MISTRAL-7B-SACHI: French-Moore Translation Model},author = {Salif SAWADOGO at BurkimbIA},year = {2025},url = {https://huggingface.co/burkimbia/BIA-MISTRAL-7B-SACHI}}