Instructions to use hyper-accel/tiny-random-minimax with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hyper-accel/tiny-random-minimax with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="hyper-accel/tiny-random-minimax")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hyper-accel/tiny-random-minimax")
model = AutoModelForCausalLM.from_pretrained("hyper-accel/tiny-random-minimax")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hyper-accel/tiny-random-minimax with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hyper-accel/tiny-random-minimax"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hyper-accel/tiny-random-minimax",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hyper-accel/tiny-random-minimax

SGLang

How to use hyper-accel/tiny-random-minimax with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "hyper-accel/tiny-random-minimax" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hyper-accel/tiny-random-minimax",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "hyper-accel/tiny-random-minimax" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hyper-accel/tiny-random-minimax",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use hyper-accel/tiny-random-minimax with Docker Model Runner:
```
docker model run hf.co/hyper-accel/tiny-random-minimax
```

tiny-random-minimax / config.json

HyperAccel

Upload tiny-random minimax model

0494ccb verified about 2 months ago

raw

history blame

1.84 kB

	{
	"_attn_implementation_autoset": false,
	"add_cross_attention": false,
	"architectures": [
	"MiniMaxForCausalLM"
	],
	"attention_dropout": 0.0,
	"block_size": 256,
	"bos_token_id": null,
	"cross_attention_hidden_size": null,
	"decoder_start_token_id": null,
	"dtype": "bfloat16",
	"eos_token_id": null,
	"finetuning_task": null,
	"full_attn_alpha_factor": 3.5565588200778455,
	"full_attn_beta_factor": 1.0,
	"head_dim": 64,
	"hidden_act": "silu",
	"hidden_size": 512,
	"initializer_range": 0.02,
	"intermediate_size": 256,
	"is_decoder": false,
	"layer_types": [
	"linear_attention",
	"linear_attention",
	"linear_attention",
	"linear_attention",
	"linear_attention",
	"linear_attention",
	"linear_attention",
	"full_attention"
	],
	"linear_attn_alpha_factor": 3.5565588200778455,
	"linear_attn_beta_factor": 1.0,
	"max_position_embeddings": 10240000,
	"mlp_alpha_factor": 3.5565588200778455,
	"mlp_beta_factor": 1.0,
	"model_type": "minimax",
	"num_attention_heads": 8,
	"num_experts_per_tok": 2,
	"num_hidden_layers": 8,
	"num_key_value_heads": 2,
	"num_local_experts": 4,
	"output_router_logits": false,
	"pad_token_id": null,
	"postnorm": true,
	"prefix": null,
	"pruned_heads": {},
	"rms_norm_eps": 1e-05,
	"rope_parameters": {
	"rope_theta": 10000000,
	"rope_type": "default"
	},
	"rotary_dim": 32,
	"router_aux_loss_coef": 0.001,
	"router_jitter_noise": 0.0,
	"sep_token_id": null,
	"shared_intermediate_size": 0,
	"shared_moe_mode": "sigmoid",
	"sliding_window": null,
	"task_specific_params": null,
	"tf_legacy_loss": false,
	"tie_encoder_decoder": false,
	"tie_word_embeddings": false,
	"tokenizer_class": null,
	"torchscript": false,
	"transformers_version": "5.3.0",
	"use_bfloat16": false,
	"use_cache": true,
	"vocab_size": 200064
	}