Instructions to use Bluebear7878/eqaq-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Bluebear7878/eqaq-v2 with Transformers:

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

pipe = pipeline("text-generation", model="Bluebear7878/eqaq-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Bluebear7878/eqaq-v2")
model = AutoModelForCausalLM.from_pretrained("Bluebear7878/eqaq-v2", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Bluebear7878/eqaq-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Bluebear7878/eqaq-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Bluebear7878/eqaq-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Bluebear7878/eqaq-v2

SGLang

How to use Bluebear7878/eqaq-v2 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 "Bluebear7878/eqaq-v2" \
    --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": "Bluebear7878/eqaq-v2",
		"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 "Bluebear7878/eqaq-v2" \
        --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": "Bluebear7878/eqaq-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Bluebear7878/eqaq-v2 with Docker Model Runner:
```
docker model run hf.co/Bluebear7878/eqaq-v2
```

mg commited on May 23

Commit

effeaeb

verified ·

1 Parent(s): 4622bf4

Document expected EQAQ v2 performance

Browse files

Files changed (1) hide show

README.md +50 -0

README.md CHANGED Viewed

@@ -34,6 +34,56 @@ Repository layout:
 The root model is the target model. The draft directories are EAGLE3 draft
 models for SGLang speculative decoding and are not standalone target models.
 Expected SGLang usage shape:
 ```bash

 The root model is the target model. The draft directories are EAGLE3 draft
 models for SGLang speculative decoding and are not standalone target models.
+## Expected Performance
+These numbers are local measurements from the EQC competition protocol harness,
+not an official leaderboard score. The official submission uploaded
+successfully, but the evaluation job failed before scoring because the service
+could not provision the requested ML compute capacity.
+Recommended route setup for the measured run:
+- Target model: repository root AWQ model
+- Latency, MMLU-Pro, IFEval draft: `drafts/q028-fast-sglangcompat`
+- GPQA/thinking draft: `drafts/q004-chatthink-sglangcompat`
+- SGLang speculative decoding: EAGLE3, `speculative-num-steps=10`,
+  `speculative-eagle-topk=2`, `speculative-num-draft-tokens=20`
+### Local latency
+Measured with the EQC latency protocol: `/v1/completions`, logical batch size
+1, 5 warmup runs, 50 measurement runs per category.
+| Category | Prompt / new tokens | Median latency | Baseline | Speedup |
+|---|---:|---:|---:|---:|
+| short | 64 / 128 | 228.87 ms | 2582 ms | 11.28x |
+| medium | 2048 / 256 | 475.62 ms | 5441 ms | 11.44x |
+| long | 8192 / 256 | 847.43 ms | 6576 ms | 7.76x |
+Average local speedup was **9.41x** against the protocol baseline. A
+submission-aligned smoke run with a more conservative single-image setup
+measured about **4.39x** average speedup over 3 runs per category.
+Observed speculative accept rate in the active local SGLang run was low,
+roughly **6%** over recent decode batches, so the latency gain should be
+understood as the combined effect of SGLang serving settings, CUDA graph, and
+speculative decoding rather than high draft acceptance alone.
+### Local quality
+Measured in the same local full protocol run:
+| Benchmark | Metric | Score | Gate |
+|---|---|---:|---:|
+| MMLU-Pro | exact_match, custom-extract | 0.6525 | 0.621 |
+| IFEval | inst_level_strict_acc | 0.8106 | 0.814 |
+| GPQA-Diamond | exact_match, flexible-extract | 0.4293 | 0.630 |
+The local run passed the latency gate and MMLU-Pro, but did **not** pass the
+full quality gate because IFEval was slightly below threshold and GPQA-Diamond
+was substantially below threshold. Treat this package as a speed-oriented EQC
+artifact, not a confirmed quality-passing competition submission.
 Expected SGLang usage shape:
 ```bash