Instructions to use Eugleo/climbmix-d26-10tpp-noright with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Eugleo/climbmix-d26-10tpp-noright with Transformers:

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

pipe = pipeline("text-generation", model="Eugleo/climbmix-d26-10tpp-noright", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Eugleo/climbmix-d26-10tpp-noright", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Eugleo/climbmix-d26-10tpp-noright with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Eugleo/climbmix-d26-10tpp-noright"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Eugleo/climbmix-d26-10tpp-noright",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Eugleo/climbmix-d26-10tpp-noright

SGLang

How to use Eugleo/climbmix-d26-10tpp-noright 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 "Eugleo/climbmix-d26-10tpp-noright" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Eugleo/climbmix-d26-10tpp-noright",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Eugleo/climbmix-d26-10tpp-noright" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Eugleo/climbmix-d26-10tpp-noright",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Eugleo/climbmix-d26-10tpp-noright with Docker Model Runner:
```
docker model run hf.co/Eugleo/climbmix-d26-10tpp-noright
```

climbmix-d26-10tpp-noright / config.json

Eugleo

exp-088 noright: d26 10-TPP base model on ClimbMix with the noright filter (arm d26-r10-a737eac7df78)

970f21f verified 21 days ago

Raw History Blame Contribute Delete

977 Bytes

	{
	"architectures": [
	"NanochatGPTForCausalLM"
	],
	"auto_map": {
	"AutoConfig": "configuration_nanochat_gpt.NanochatGPTConfig",
	"AutoModel": "modeling_nanochat_gpt.NanochatGPTModel",
	"AutoModelForCausalLM": "modeling_nanochat_gpt.NanochatGPTForCausalLM"
	},
	"backout_layer": null,
	"bos_token_id": 32759,
	"dtype": "bfloat16",
	"eos_token_id": 32759,
	"final_logit_softcapping": 15.0,
	"hidden_size": 1664,
	"intermediate_size": 6656,
	"logit_softcap": 15.0,
	"max_position_embeddings": 2048,
	"model_type": "nanochat_gpt",
	"num_attention_heads": 13,
	"num_hidden_layers": 26,
	"num_key_value_heads": 13,
	"qk_sharpen_scale": null,
	"rope_theta": 100000.0,
	"smear_gate_channels": 24,
	"tie_word_embeddings": false,
	"transformers_version": "5.14.1",
	"use_resid_lambdas": false,
	"use_smear": false,
	"use_x0_lambdas": false,
	"value_embedding_layers": [],
	"ve_gate_channels": 12,
	"vocab_size": 32768,
	"window_pattern": "L"
	}