Instructions to use feeltheAGI/Maverick-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use feeltheAGI/Maverick-7B with Transformers:

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

pipe = pipeline("text-generation", model="feeltheAGI/Maverick-7B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("feeltheAGI/Maverick-7B")
model = AutoModelForCausalLM.from_pretrained("feeltheAGI/Maverick-7B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use feeltheAGI/Maverick-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "feeltheAGI/Maverick-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "feeltheAGI/Maverick-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/feeltheAGI/Maverick-7B

SGLang

How to use feeltheAGI/Maverick-7B 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 "feeltheAGI/Maverick-7B" \
    --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": "feeltheAGI/Maverick-7B",
		"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 "feeltheAGI/Maverick-7B" \
        --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": "feeltheAGI/Maverick-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use feeltheAGI/Maverick-7B with Docker Model Runner:
```
docker model run hf.co/feeltheAGI/Maverick-7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Maverick-7B

This model is a merge of the following models:

🏆 Evaluation

TruthfulQA

Task	Version	Metric	Value		Stderr
truthfulqa_mc	1	mc1	0.5165	±	0.0175
		mc2	0.6661	±	0.0152

GPT4ALL

Task	Version	Metric	Value		Stderr
arc_challenge	0	acc	0.6442	±	0.0140
		acc_norm	0.6570	±	0.0139
arc_easy	0	acc	0.8645	±	0.0070
		acc_norm	0.8304	±	0.0077
boolq	1	acc	0.8850	±	0.0056
hellaswag	0	acc	0.6813	±	0.0047
		acc_norm	0.8571	±	0.0035
openbookqa	0	acc	0.3640	±	0.0215
		acc_norm	0.4800	±	0.0224
piqa	0	acc	0.8324	±	0.0087
		acc_norm	0.8460	±	0.0084
winogrande	0	acc	0.7869	±	0.0115

AGIEval

Task	Version	Metric	Value		Stderr
agieval_aqua_rat	0	acc	0.2717	±	0.0280
		acc_norm	0.2559	±	0.0274
agieval_logiqa_en	0	acc	0.3902	±	0.0191
		acc_norm	0.3856	±	0.0191
agieval_lsat_ar	0	acc	0.2565	±	0.0289
		acc_norm	0.2478	±	0.0285
agieval_lsat_lr	0	acc	0.5118	±	0.0222
		acc_norm	0.5216	±	0.0221
agieval_lsat_rc	0	acc	0.6543	±	0.0291
		acc_norm	0.6506	±	0.0291
agieval_sat_en	0	acc	0.7961	±	0.0281
		acc_norm	0.8010	±	0.0279
agieval_sat_en_without_passage	0	acc	0.4660	±	0.0348
		acc_norm	0.4709	±	0.0349
agieval_sat_math	0	acc	0.3227	±	0.0316
		acc_norm	0.3045	±	0.0311

Bigbench

Task	Version	Metric	Value		Stderr
bigbench_causal_judgement	0	multiple_choice_grade	0.5684	±	0.0360
bigbench_date_understanding	0	multiple_choice_grade	0.6612	±	0.0247
bigbench_disambiguation_qa	0	multiple_choice_grade	0.4380	±	0.0309
bigbench_geometric_shapes	0	multiple_choice_grade	0.2173	±	0.0218
		exact_str_match	0.0000	±	0.0000
bigbench_logical_deduction_five_objects	0	multiple_choice_grade	0.3320	±	0.0211
bigbench_logical_deduction_seven_objects	0	multiple_choice_grade	0.2243	±	0.0158
bigbench_logical_deduction_three_objects	0	multiple_choice_grade	0.5667	±	0.0287
bigbench_movie_recommendation	0	multiple_choice_grade	0.4260	±	0.0221
bigbench_navigate	0	multiple_choice_grade	0.5310	±	0.0158
bigbench_reasoning_about_colored_objects	0	multiple_choice_grade	0.7230	±	0.0100
bigbench_ruin_names	0	multiple_choice_grade	0.5379	±	0.0236
bigbench_salient_translation_error_detection	0	multiple_choice_grade	0.2956	±	0.0145
bigbench_snarks	0	multiple_choice_grade	0.6961	±	0.0343
bigbench_sports_understanding	0	multiple_choice_grade	0.7424	±	0.0139
bigbench_temporal_sequences	0	multiple_choice_grade	0.4690	±	0.0158
bigbench_tracking_shuffled_objects_five_objects	0	multiple_choice_grade	0.2304	±	0.0119
bigbench_tracking_shuffled_objects_seven_objects	0	multiple_choice_grade	0.1880	±	0.0093
bigbench_tracking_shuffled_objects_three_objects	0	multiple_choice_grade	0.5667	±	0.0287

Downloads last month: 84

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for feeltheAGI/Maverick-7B

Quantizations

3 models