Skip to content

docs(readme): add openinfer (Rust) as a DFlash backend #141

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
xiaguan wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs(readme): add openinfer (Rust) as a DFlash backend #141

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,6 +160,18 @@ for r in stream_generate(model, draft, tokenizer, prompt, block_size=16, max_tok
print(f"\nThroughput: {tps:.2f} tok/s")
```

### openinfer (Rust)

[openinfer](https://github.com/openinfer-project/openinfer) is a pure Rust + CUDA inference engine (no PyTorch) with native DFlash support for Qwen3-4B / 8B, using the [`z-lab/Qwen3-4B-DFlash-b16`](https://huggingface.co/z-lab/Qwen3-4B-DFlash-b16) and [`z-lab/Qwen3-8B-DFlash-b16`](https://huggingface.co/z-lab/Qwen3-8B-DFlash-b16) drafters.

```bash
cargo run --release -- \
--model-path models/Qwen3-4B \
--dflash-draft-model-path models/Qwen3-4B-DFlash-b16
```

Single-stream decode speedup: **1.82× on RTX 5070 Ti**, **1.56× on RTX 5090**.

## 📊 Evaluation

All benchmarks share the same datasets (gsm8k, math500, humaneval, mbpp, mt-bench). Datasets are automatically downloaded and cached as JSONL in `cache/` on first run.
Expand Down