tairov/llama2.mojo
Inference Llama 2 in one file of pure 🔥
Total stars: 280
Stars trend:
11 Sep 2023
12 Sep 2023
#inference, #llama, #llama2, #modular, #mojo, #parallelize, #performance, #simd, #tensor, #vectorization
Inference Llama 2 in one file of pure 🔥
Total stars: 280
Stars trend:
11 Sep 2023
10pm ██▌ +20
11pm █▌ +12
12 Sep 2023
12am █▎ +10
1am ██▍ +19
2am ██ +16
3am █▉ +15
4am ██ +16
5am ███▏ +25
6am ███▌ +28
7am ██ +16
8am ███▏ +25
9am ███▋ +29
#inference, #llama, #llama2, #modular, #mojo, #parallelize, #performance, #simd, #tensor, #vectorization
🔥1
ashvardanian/SimSIMD
Vector Similarity Functions 3x-200x Faster than SciPy and NumPy — for Python, JavaScript, and C 11, supporting f64, f32, f16, i8, and binary vectors using SIMD for both x86 AVX2 & AVX-512 and Arm NEON & SVE 📐
Language:C
Total stars: 397
Stars trend:
#c
#arm, #armneon, #armsve, #assembly, #avx2, #avx512, #distancecalculation, #distancemeasures, #informationretrieval, #metrics, #neon, #numpy, #python, #scipy, #simd, #simdinstructions, #similaritymeasures, #similaritysearch, #vectorsearch
Vector Similarity Functions 3x-200x Faster than SciPy and NumPy — for Python, JavaScript, and C 11, supporting f64, f32, f16, i8, and binary vectors using SIMD for both x86 AVX2 & AVX-512 and Arm NEON & SVE 📐
Language:C
Total stars: 397
Stars trend:
23 Jan 2024
11pm ▏ +1
24 Jan 2024
12am ▋ +5
1am ███ +24
2am ███ +24
3am ███▋ +29
4am █▉ +15
5am █▍ +11
6am █▎ +10
7am █▎ +10
8am ██ +16
9am █▋ +13#c
#arm, #armneon, #armsve, #assembly, #avx2, #avx512, #distancecalculation, #distancemeasures, #informationretrieval, #metrics, #neon, #numpy, #python, #scipy, #simd, #simdinstructions, #similaritymeasures, #similaritysearch, #vectorsearch
timeplus-io/proton
A streaming SQL engine, a fast and lightweight alternative to Apache Flink, 🚀 powered by ClickHouse.
Language:C++
Total stars: 596
Stars trend:
#cplusplus
#analytics, #clickhouse, #confluent, #cpp, #flinkalternative, #highperformance, #kakfa, #ksqldbalternative, #redpanda, #simd, #singlebinary, #sql, #streamprocessing, #streamingsql, #udf
A streaming SQL engine, a fast and lightweight alternative to Apache Flink, 🚀 powered by ClickHouse.
Language:C++
Total stars: 596
Stars trend:
30 Jan 2024
10pm ▏ +1
11pm ███▉ +31
31 Jan 2024
12am ███▏ +25
1am ███▏ +25
2am █▋ +13
3am █▌ +12
4am █▏ +9
5am █ +8
6am █▌ +12
7am █ +8
8am █▎ +10#cplusplus
#analytics, #clickhouse, #confluent, #cpp, #flinkalternative, #highperformance, #kakfa, #ksqldbalternative, #redpanda, #simd, #singlebinary, #sql, #streamprocessing, #streamingsql, #udf
ashvardanian/StringZilla
Up to 10x faster strings for C, C++, Python, Rust, and Swift, leveraging SWAR and SIMD on Arm Neon and x86 AVX2 & AVX-512-capable chips to accelerate search, sort, edit distances, alignment scores, etc 🦖
Language:C++
Total stars: 1230
Stars trend:
#cplusplus
#beautifulsoup, #commoncrawl, #csv, #dataset, #html, #informationretrieval, #json, #laion, #ndjson, #parser, #patternrecognition, #simd, #sortingalgorithms, #string, #stringmanipulation, #stringmatching, #stringparsing, #stringsearch, #substring
Up to 10x faster strings for C, C++, Python, Rust, and Swift, leveraging SWAR and SIMD on Arm Neon and x86 AVX2 & AVX-512-capable chips to accelerate search, sort, edit distances, alignment scores, etc 🦖
Language:C++
Total stars: 1230
Stars trend:
24 Feb 2024
8am ▊ +6
9am ▍ +3
10am ▋ +5
11am █▏ +9
12pm ▊ +6
1pm ▉ +7
2pm █ +8
3pm █▏ +9
4pm █▏ +9
5pm █ +8
6pm ▋ +5
7pm ▍ +3#cplusplus
#beautifulsoup, #commoncrawl, #csv, #dataset, #html, #informationretrieval, #json, #laion, #ndjson, #parser, #patternrecognition, #simd, #sortingalgorithms, #string, #stringmanipulation, #stringmatching, #stringparsing, #stringsearch, #substring
google/highway
Performance-portable, length-agnostic SIMD with runtime dispatch
Language:C++
Total stars: 3318
Stars trend:
#cplusplus
#avx, #avx512, #avxinstructions, #avx2, #avx512, #intrinsics, #neon, #simd, #simdinstructions, #simdintrinsics, #simdlibrary, #simdparallelism, #simdprogramming, #sse42, #wasm
Performance-portable, length-agnostic SIMD with runtime dispatch
Language:C++
Total stars: 3318
Stars trend:
25 Feb 2024
6pm ▍ +3
7pm ▏ +1
8pm ▎ +2
9pm ▏ +1
10pm ▌ +4
11pm ▋ +5
26 Feb 2024
12am ▊ +6
1am █▎ +10
2am ▋ +5
3am ▌ +4
4am ▍ +3
5am █▎ +10#cplusplus
#avx, #avx512, #avxinstructions, #avx2, #avx512, #intrinsics, #neon, #simd, #simdinstructions, #simdintrinsics, #simdlibrary, #simdparallelism, #simdprogramming, #sse42, #wasm
ogxd/gxhash
The fastest hashing algorithm 📈
Language:Rust
Total stars: 330
Stars trend:
#rust
#cryptography, #hash, #hashing, #hashmap, #ilp, #nostd, #performance, #simd
The fastest hashing algorithm 📈
Language:Rust
Total stars: 330
Stars trend:
13 May 2024
4pm ▏ +1
5pm ▋ +5
6pm █▋ +13
7pm █▉ +15
8pm ██▊ +22
9pm ▉ +7
10pm ▊ +6
11pm ▍ +3
14 May 2024
12am ▌ +4#rust
#cryptography, #hash, #hashing, #hashmap, #ilp, #nostd, #performance, #simd
kelindar/search
Go library for embedded vector search and semantic embeddings using llama.cpp
Language:Go
Total stars: 118
Stars trend:
#go
#ai, #bert, #embeddings, #gguf, #gpu, #llamacpp, #searchengine, #semanticsearch, #simd, #vectorsearch
Go library for embedded vector search and semantic embeddings using llama.cpp
Language:Go
Total stars: 118
Stars trend:
29 Oct 2024
10pm ▌ +4
11pm █▉ +15
30 Oct 2024
12am ██ +16
1am ██▊ +22
2am █▎ +10
3am █ +8#go
#ai, #bert, #embeddings, #gguf, #gpu, #llamacpp, #searchengine, #semanticsearch, #simd, #vectorsearch
questdb/questdb
QuestDB is a high performance, open-source, time-series database
Language:Java
Total stars: 15503
Stars trend:
#java
#capitalmarkets, #cpp, #database, #financialanalysis, #grafana, #java, #lowlatency, #marketdata, #olap, #postgresql, #questdb, #realtimeanalytics, #sensordata, #simd, #sql, #tickdata, #timeseries, #timeseriesdatabase, #tsdb
QuestDB is a high performance, open-source, time-series database
Language:Java
Total stars: 15503
Stars trend:
6 Jul 2025
6am ████▍ +35
7am ███▏ +25
8am █▎ +10
9am ██▏ +17
10am █▋ +13#java
#capitalmarkets, #cpp, #database, #financialanalysis, #grafana, #java, #lowlatency, #marketdata, #olap, #postgresql, #questdb, #realtimeanalytics, #sensordata, #simd, #sql, #tickdata, #timeseries, #timeseriesdatabase, #tsdb
simdjson/simdjson
Parsing gigabytes of JSON per second : used by Facebook/Meta Velox, the Node.js runtime, ClickHouse, WatermelonDB, Apache Doris, Milvus, StarRocks
Language:C++
Total stars: 21788
Stars trend:
#cplusplus
#aarch64, #arm, #arm64, #avx2, #avx512, #cplusplus, #clang, #clangcl, #cpp11, #gcccompiler, #json, #jsonparser, #jsonpointer, #loongarch, #neon, #simd, #sse42, #vs2019, #x64
Parsing gigabytes of JSON per second : used by Facebook/Meta Velox, the Node.js runtime, ClickHouse, WatermelonDB, Apache Doris, Milvus, StarRocks
Language:C++
Total stars: 21788
Stars trend:
14 Sep 2025
1am ▊ +6
2am ▊ +6
3am ▊ +6
4am █▋ +13
5am █▋ +13
6am █ +8
7am █▌ +12
8am █▏ +9
9am ▌ +4
10am █▎ +10
11am █▌ +12
12pm █ +8#cplusplus
#aarch64, #arm, #arm64, #avx2, #avx512, #cplusplus, #clang, #clangcl, #cpp11, #gcccompiler, #json, #jsonparser, #jsonpointer, #loongarch, #neon, #simd, #sse42, #vs2019, #x64
❤1
RyanCodrai/turbovec
A vector index built on TurboQuant, written in Rust with Python bindings
Language:Python
Total stars: 2066
Stars trend:
#python
#ann, #avx512, #embedding, #embeddings, #faiss, #nearestneighbor, #neon, #python, #quant, #quantization, #rag, #rust, #simd, #turboquant, #vectorsearch
A vector index built on TurboQuant, written in Rust with Python bindings
Language:Python
Total stars: 2066
Stars trend:
21 May 2026
4pm ▏ +1
5pm +0
6pm +0
7pm ▏ +1
8pm ▍ +3
9pm ▎ +2
10pm ▍ +3
11pm ▉ +7
22 May 2026
12am ▊ +6
1am █ +8#python
#ann, #avx512, #embedding, #embeddings, #faiss, #nearestneighbor, #neon, #python, #quant, #quantization, #rag, #rust, #simd, #turboquant, #vectorsearch
FareedKhan-dev/kimi-k3-in-c
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
Language:C
Total stars: 3210
#c
#avx2, #c99, #cpuinference, #deeplearning, #fromscratch, #inferenceengine, #kimik3, #linearattention, #llm, #llminference, #machinelearning, #memoryefficient, #mixtureofexperts, #moe, #mxfp4, #quantization, #simd, #systemsprogramming, #transformer, #zerodependencies
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
Language:C
Total stars: 3210
#c
#avx2, #c99, #cpuinference, #deeplearning, #fromscratch, #inferenceengine, #kimik3, #linearattention, #llm, #llminference, #machinelearning, #memoryefficient, #mixtureofexperts, #moe, #mxfp4, #quantization, #simd, #systemsprogramming, #transformer, #zerodependencies