GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Prompt48
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Prompt48
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
| Qwen3-8B (base) | This model | |
|---|---|---|
| Playable, valid music | 1/160 (0.6%) | 19/160 (11.9%) |
| Prompts with ≥1 valid piece | 1/40 | 13/40 |
| Mean grader score | 0.002 | 0.041 |
| Avg answer length | 2171 tokens | 1256 tokens |
~20× more valid music after 30 GRPO steps (~1.7 h on 1× H100). Still only ~12% valid — a starting point, not a finished composer.
lora_adapter/.Use long, detailed composition briefs like the training data (short prompts give short, repetitive output). Prompt for a piece and ask for ABC in a ```abc block. Disable thinking:
python
from transformers import AutoModelForCausalLM, AutoTokenizerm = "Prompt48/Qwen3-8B-MiMo-Music-GRPO"tok = AutoTokenizer.from_pretrained(m); model = AutoModelForCausalLM.from_pretrained(m, torch_dtype="auto", device_map="auto")msgs = [{"role": "system", "content": "You are a composer. Reply with one complete piece in ABC notation inside a ```abc code block. Start with X:1 and include T:, M:, L:, Q:, K: headers."}, {"role": "user", "content": "Compose a Viennese waltz in Eb major, 181 BPM allegro, 3/4, about 96 bars and about three and a half minutes, scored for strings and harp in 4 voices. Texture: a slightly anticipated waltz bass under an elegant sweeping tune. Open with 8 bars of strings alone, then layer the other parts in one at a time. The main section runs on a 4-bar diatonic loop of I–IV–V–I, unchanged throughout. Thin the middle section to two voices for two descending phrases, then return with everything for the fullest passage. Dynamics start at p, push to mf in the middle and settle back to p for a sustained tonic close. Aim for the feeling of a rooftop above neon. Use no ornament symbols; write every note out. Output the complete piece as ABC notation."}]ids = tok.apply_chat_template(msgs, add_generation_prompt=True, enable_thinking=False, return_tensors="pt").to(model.device)print(tok.decode(model.generate(ids, max_new_tokens=2048, do_sample=True, temperature=1.0)[0][ids.shape[1]:]))Render: abc2midi piece.abc -o piece.mid, then any MIDI player / fluidsynth.
GGUF (Ollama / LM Studio) — re-checked on the same test set: merged bf16 14.4% valid, Q8_0 18.8%, Q4_K_M 5.0% (use Q8_0): Prompt48/Qwen3-8B-MiMo-Music-GRPO-GGUF
Xiaomi MiMo team (environment, grader, method), Qwen team (base model), TRL / vLLM.
from transformers import AutoModelForCausalLM, AutoTokenizerm = "Prompt48/Qwen3-8B-MiMo-Music-GRPO"tok = AutoTokenizer.from_pretrained(m); model = AutoModelForCausalLM.from_pretrained(m, torch_dtype="auto", device_map="auto")msgs = [{"role": "system", "content": "You are a composer. Reply with one complete piece in ABC notation inside a ```abc code block. Start with X:1 and include T:, M:, L:, Q:, K: headers."}, {"role": "user", "content": "Compose a Viennese waltz in Eb major, 181 BPM allegro, 3/4, about 96 bars and about three and a half minutes, scored for strings and harp in 4 voices. Texture: a slightly anticipated waltz bass under an elegant sweeping tune. Open with 8 bars of strings alone, then layer the other parts in one at a time. The main section runs on a 4-bar diatonic loop of I–IV–V–I, unchanged throughout. Thin the middle section to two voices for two descending phrases, then return with everything for the fullest passage. Dynamics start at p, push to mf in the middle and settle back to p for a sustained tonic close. Aim for the feeling of a rooftop above neon. Use no ornament symbols; write every note out. Output the complete piece as ABC notation."}]ids = tok.apply_chat_template(msgs, add_generation_prompt=True, enable_thinking=False, return_tensors="pt").to(model.device)print(tok.decode(model.generate(ids, max_new_tokens=2048, do_sample=True, temperature=1.0)[0][ids.shape[1]:]))