qdrant/fastembed

View on GitHub

Fast, Accurate, Lightweight Python library to make State of the Art Embedding

Stars: 3.2k
Forks: 248
Open beginner issues: 0
Indexed issues: 78
Avg merge: 4d 8h
Merged PRs (30d): 4

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 16, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

78 open issues indexed

[Bug]: Apache License 2.0 in pypi.org Open

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

qdrant/fastembed#712 · 1 comment · Opened Sep 19, 2026
[Bug]: gte-base mixed-length batches fail with fixed padding in 0.8.0 Open

Difficulty 3/5 1-2 days Newbie friendliness 76/100

qdrant/fastembed#703 · Opened Sep 3, 2026
ibm-granite/granite-embedding-small-english-r2 Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

qdrant/fastembed#701 · Opened Sep 3, 2026
[Bug]: tokenizer_config max_length caps several models at 128 tokens Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

qdrant/fastembed#689 · 2 comments · Opened Aug 24, 2026
Silent all-NaN embeddings when a model file is incompatible with the runtime (0.8.0 still ships the fp16 jina-de model fixed by #623) Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

qdrant/fastembed#688 · 1 comment · Opened Aug 24, 2026
[Feature]: make config.json and special_tokens_map.json optional Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

qdrant/fastembed#686 · 1 comment · Opened Aug 20, 2026
[Bug]: models with max_length=0 in tokenizer config break Open

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

qdrant/fastembed#685 · Opened Aug 19, 2026
Resolving License Issues Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#679 · 1 comment · Opened Aug 18, 2026
Qdrant/bm25 refetches from the Hub when fully cached (offline probe ignores the mock model_file) Open

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

qdrant/fastembed#675 · Opened Aug 14, 2026
[Bug]: Closing a parallel result iterator early hangs while joining workers Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

qdrant/fastembed#669 · Opened Aug 11, 2026
[Feature]: Support pluggable stemmer for BM25 (enable Polish and other non‑Snowball languages) Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

qdrant/fastembed#654 · Opened Jul 28, 2026
[Bug]: Resize transform passes non-square dimensions to Pillow in the wrong order Open

Difficulty 2/5 1-3 hours Newbie friendliness 76/100

qdrant/fastembed#649 · Opened Jul 3, 2026
[Model]: LateOn Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#641 · Opened May 31, 2026
Hard to get the full model name of long models like sentence-transformers paraphrase Open

Difficulty 2/5 1-3 hours Newbie friendliness 50/100

qdrant/fastembed#632 · 1 comment · Opened May 19, 2026
[Model]: s2593817/sft-sql-embedding Open

Difficulty 3/5 1-2 days Newbie friendliness 48/100

qdrant/fastembed#628 · Opened Apr 14, 2026
[Bug]: Tar path traversal (Zip Slip) in decompress_to_cache — arbitrary file write outside cache directory Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

qdrant/fastembed#626 · Opened Apr 11, 2026
[Bug]: Segmentation Fault or AssertionError during initialization on Python 3.14.2 Open

Difficulty 4/5 3-5 days Newbie friendliness 42/100

qdrant/fastembed#618 · 4 comments · 1 reaction · Opened Mar 26, 2026
[Model]: perplexity-ai/pplx-embed-v1-0.6b Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#613 · 1 reaction · Opened Mar 15, 2026
[Feature]: Support Chinese Tokenization for BM25 using Jieba Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

qdrant/fastembed#610 · 2 comments · Opened Mar 6, 2026
GPU users: onnxruntime (CPU) overwrites onnxruntime-gpu binaries when both are installed by pip/uv Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#608 · Opened Feb 25, 2026
[Model]: jina v5 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#607 · 5 reactions · Opened Feb 25, 2026
[Feature]: RFC: Biological embeddings support Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

qdrant/fastembed#597 · 1 reaction · Opened Jan 12, 2026
[Feature]: support qwen3-vl Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

qdrant/fastembed#596 · 3 reactions · Opened Jan 10, 2026
[Feature]: Add custom token corpus for other languages for bm25 Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

qdrant/fastembed#594 · 1 comment · 2 reactions · Opened Dec 19, 2025
[Feature] Support nomic-ai/CodeRankEmbed model Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

qdrant/fastembed#587 · 1 reaction · Opened Dec 5, 2025
[Feature]: make token count accessible without loading the model Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

qdrant/fastembed#584 · Opened Dec 1, 2025
[Feature]: add docs for exposed onnx session options, cache_dir and local_files_only Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

qdrant/fastembed#579 · 1 comment · Opened Nov 25, 2025
[Feature]: Add python3.14 support Open

Difficulty 3/5 1-2 days Newbie friendliness 42/100

qdrant/fastembed#576 · 7 comments · 2 reactions · Opened Nov 14, 2025
[Feature]: Add CLI to launch FastEmbed as an Embedding API service (OpenAI-compatible) Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

qdrant/fastembed#571 · 2 comments · 3 reactions · Opened Nov 5, 2025
[Bug]: downloaded embedding models should not go to a temporary folder by default Open

Difficulty 3/5 1-2 days Newbie friendliness 48/100

qdrant/fastembed#569 · 3 comments · Opened Oct 31, 2025
Sparse Embeddings (BM25) and Persian Language Support Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

qdrant/fastembed#558 · 2 comments · 2 reactions · Opened Aug 28, 2025
Does converting model to onnx format cause performance degradation of model? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

qdrant/fastembed#554 · 5 comments · Opened Aug 20, 2025
how could fastembed load the local model? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

qdrant/fastembed#549 · 4 comments · 1 reaction · Opened Aug 7, 2025
[Bug]: SparseTextEmbedding needs to implement aembed for non-blocking use (async/await) Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

qdrant/fastembed#546 · 2 comments · Opened Jul 24, 2025
[Feature]: miniCOIL: no way to control input sequence length (and VRAM usage) natively Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#544 · 1 comment · Opened Jul 16, 2025
[Performance]: 🐢 Sparse Embedding with SPLADE is Very Slow + OOM For larger batches - Expected Behavior? Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

qdrant/fastembed#539 · 2 comments · Opened Jun 29, 2025
[Model]: granite-embedding-107m-multilingual Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#538 · Opened Jun 24, 2025
[Bug]: wrong retriever from version 0.6.0 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#536 · 1 comment · Opened Jun 19, 2025
[Question] on an M2 Max FastEmbed is way slower than SentenceTransformers Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

qdrant/fastembed#535 · 6 comments · 1 reaction · Opened Jun 19, 2025
[Feature]: HF source for custom models should allow local directory Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

qdrant/fastembed#532 · 1 comment · Opened Jun 14, 2025
[Bug]: Docstring does not align with behaviour Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

qdrant/fastembed#531 · 3 comments · Opened Jun 11, 2025
[Feature]: specify output to use for custom models Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#530 · Opened Jun 7, 2025
[Feature]: Last token pooling for causal embedding models Open

Difficulty 3/5 1-2 days Newbie friendliness 42/100

qdrant/fastembed#529 · Opened Jun 7, 2025
[Model]: `Qwen/Qwen3-Embedding-0.6B-GGUF` Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

qdrant/fastembed#528 · 4 comments · 5 reactions · Opened Jun 5, 2025
[Model]: jinaai/jina-clip-v2 Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

qdrant/fastembed#509 · 1 reaction · Opened Apr 22, 2025
[Feature]: How to add self-defined emb model OR ollama emb Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

qdrant/fastembed#507 · Opened Apr 17, 2025
[Feature]: support HF_ENDPOINT environment variable Open

Difficulty 3/5 1-2 days Newbie friendliness 52/100

qdrant/fastembed#502 · 2 comments · Opened Mar 31, 2025
[Model]: BAAI/bge-reranker-v2-m3 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

qdrant/fastembed#494 · Opened Mar 6, 2025
[Feature]: Multioutput models Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

qdrant/fastembed#485 · Opened Feb 21, 2025
ONNXRuntime taking up too much memory Open

qdrant/fastembed#483 · 6 comments · 1 assignee · Opened Feb 20, 2025

qdrant/fastembed

Get new issues in your inbox