Ecosyste.ms: OpenCollective

An open API service for software projects hosted on Open Collective.

github.com/radareorg/r2ai

local language model for radare2
https://github.com/radareorg/r2ai

Release 0.5.0

ce9c8da54351eafa0d1403bfff478b3267937a1d authored 11 months ago by pancake <[email protected]>
Fix missing trycatch

b493d7f59d5c9e40de57abfed44523c08837643d authored 11 months ago by pancake <[email protected]>
Ensure some logs are printed via the bulitin

bbd9be54c6f2bce53486129e3097ee99fe52f971 authored 12 months ago by pancake <[email protected]>
Support loading r2ai scripts, not just python ones

0f4a0ec40ca8e4e372245a746bf24aa28f084f6e authored 12 months ago by pancake <[email protected]>
Fix missing slurp in the repl

9aa800e96f93b3b8d264efb55496d4317f969156 authored 12 months ago by pancake <[email protected]>
Fix q4im issue

556a9c67af82c18bb3017704051496a5b6ae152b authored 12 months ago by pancake <[email protected]>
in openchatland 0106 is newer than 1210

41b418addd47b62bbdde26143b4566161e8b7b85 authored 12 months ago by pancake <[email protected]>
Add chat.code to make code block highlighting optional

ca19d77b5c7ba600f093beba6c274c8a9df7e3db authored 12 months ago by pancake <[email protected]>
Add support for the openchat model

dd1570a781c902d2a6236eb707b0a634d77ede78 authored 12 months ago by pancake <[email protected]>
Add another markdown for doc/data

540b0bb808db9168a216f79e547ac965b1cc951a authored 12 months ago by pancake <[email protected]>
Refactoring main.py into r2ai.repl for faster plugin loading times

4cc7e143231297e1b3533b8a6779ec08e3b6f6e2 authored 12 months ago by pancake <[email protected]>
Ignore hidden and package.json files from indexer

6e031fa7b3ff4a9df25774c447b834e6015c8abb authored 12 months ago by pancake <[email protected]>
next -> continue

9ce6a6ac91d00b19b283eb32dd1e054d7676ee12 authored 12 months ago by pancake <[email protected]>
Fix query length computation when chat.reply is false

01541e1c9effacd42fbc1e0b50319d64b24fc807 authored 12 months ago by pancake <[email protected]>
Improve search hint support for ferret models

9b38e51533b969c9f62a03a3128e4bb6e8ab7497 authored 12 months ago by pancake <[email protected]>
Use unique=True, batch_results="diverse") if possible in vectordb

4048ecfb07e1a7ff310c723b1668dd7e9f085f52 authored 12 months ago by pancake <[email protected]>
Use local vectordb installation (pip version is too old and buggy)

c80748188ccfb6d8e44c2ec5def82f3a762c1f5e authored 12 months ago by pancake <[email protected]>
Switch to the dolphin-phi2 model by default

7b7b66828fe44c78d5c239ec9255be92022f4de1 authored 12 months ago by pancake <[email protected]>
Add -Lj and improve -L

dd84bdd1562047c9bfc20bc3e62846164d33d5bc authored 12 months ago by pancake <[email protected]>
Add natural language compressor using the nltk module

2e5b3ebf5d2ea9da5217fca43f0a6c146ce5ddd4 authored 12 months ago by pancake <[email protected]>
Add hint for gpt-4-turbo model support

7057b5c0d710eb6e5bd7d8b5ecc695e341135685 authored about 1 year ago by pancake <[email protected]>
Add llm.maxmsglen config var

c614d85e65f5264ee119832fd75b4284aba4878a authored about 1 year ago by pancake <[email protected]>
Remove some dupped imports and fix r2 plugin integration

416654ba5d142b0517ded6b8ac6ab300138658e7 authored about 1 year ago by pancake <[email protected]>
Add support for zephyr models and improve ferret/dolphin one

089032a55c56aab0c178190c812cf65796a2ac95 authored about 1 year ago by pancake <[email protected]>
Fix chat.reply

312791b48ba83cd49d9a6b5769a696b5cdb5124f authored about 1 year ago by pancake <[email protected]>
Add support for the utopiaXL and the dolphin-PHI models

ed10c51683392a1e1cff488415740f6d8fda4e0f authored about 1 year ago by pancake <[email protected]>
Add support for the Apple's Ferret model

0c02a63411614d46c61afcac464dcffee54c5be8 authored about 1 year ago by pancake <[email protected]>
Native and python implementations can't coexist, force likewise uninstall

e730e7a102a2a3ad3f9379b9a441b7ec19b8cf2c authored about 1 year ago by pancake <[email protected]>
One less backtrace when the model is not found

f824af387c9288e4a62b82d458ef65a14e0a4a72 authored about 1 year ago by pancake <[email protected]>
Fix null deref when model cant be found in the r2ai-native

14f51a39eca527cc065c271b315ca81c1e8e9e8b authored about 1 year ago by pancake <[email protected]>
First working version of r2ai-native, still far from clean

5405e69b659dd9f16f68c6bfbf5cf37604d2fcbb authored about 1 year ago by pancake <[email protected]>
Add uninstall target

5d87f1f1743841b0e66cd1ea91402e7bcf40fbc9 authored about 1 year ago by pancake <[email protected]>
Fix build on Linux

6d2d855dcf51498ac4a2bbc0c1de7d47326cffa1 authored about 1 year ago by pancake <[email protected]>
A bit more work integrating llama with the r2ai cxx plugin

50051433afb298804fef069e50f3fcbcb464b75f authored about 1 year ago by pancake <[email protected]>
Initial rewrite of llama's main example for the native

e0b69fc07e6d4316657ea7311c223b492dddea2a authored about 1 year ago by pancake <[email protected]>
Fix -M

7579175cb9790a302ceac53036083dc3848f41fb authored about 1 year ago by pancake <[email protected]>
Initial import of the r2ai http client plugin

7891957a37206cfee3e1c7fee8ca25801610b4fc authored about 1 year ago by pancake <[email protected]>
Implement the -w command to start the http server

f51e4c33b13ddf40311fee97476fd08c9c2bb7d5 authored about 1 year ago by pancake <[email protected]>
Add the mistral 0.2 in the listing

0cbf4fd9d4943e0f6f660f6d3abe707102ceef67 authored about 1 year ago by pancake <[email protected]>
Improve error messages and fix missing import for the voice support

83708433a10ce40c08fa9627a410b81f34edefa1 authored about 1 year ago by pancake <[email protected]>
Initial import of the C rewrite of r2ai (very WIP just a PoC)

b593ceb399ec9ad95235f481dcc50827936d68b7 authored about 1 year ago by pancake <[email protected]>
Improved r2pipe with argument support

* `#!pipe r2pm -r r2ai examples/r2test.py arg1 arg2`

b69ef96c04cda8d065dfed8993eedfaaf4e399b7 authored about 1 year ago by pancake <[email protected]>
Improved support for venv and rlang command handling

d8eb08b0ef3fa6931245a4c9757f22bd5895bc0a authored about 1 year ago by pancake <[email protected]>
Initial support to index json files (WIP PoC)

95a2abe010307a7be2ff6e0c1f6cfeb0359006d6 authored about 1 year ago by pancake <[email protected]>
Vim integration through adding support to use r2ai as a hashbang to run scripts

49dbbf28ae532627f0866a4b75984964c2183c6c authored about 1 year ago by pancake <[email protected]>
Initial Wikipedia integration via `wikit`

a9f4a9d280f3b45a89762fde7841c79d6aaf843d authored about 1 year ago by pancake <[email protected]>
use ai to split relevant keywords for mastodon search

159af0d76ad0df972995534db234c9bdfdd142bc authored about 1 year ago by pancake <[email protected]>
Initial support for magicoder model and fix chat.reply

af7665f143ca8ae788d4b581fb7a2cf877d819ac authored about 1 year ago by pancake <[email protected]>
Initial support for Mastodon in vectordb

a5c6ecb38d963e6e339675ce12a5687c8e0b1f69 authored about 1 year ago by pancake <[email protected]>
Fix bug in -t

53b6273c8df509f34be643482dbf330d34b5578f authored about 1 year ago by pancake <[email protected]>
Fix -t

4b471cfa0b18a8e93d6982dab30941a5557fa603 authored about 1 year ago by pancake <[email protected]>
Fix deepseek prompt template, as it's alpaca based

84ca584e6ef25b58b211669610126322cacfd035 authored about 1 year ago by pancake <[email protected]>
Add git commit script example

c549a3ff6b0e0d662858553187998d9aec1033a4 authored about 1 year ago by pancake <[email protected]>
Add extra check when no model is selected

08ecc7930b40528a51de69b6c67a3418295de2d8 authored about 1 year ago by pancake <[email protected]>
Fix "inr2" checks, now r2ai works via r2pipe, rlang and standalone

c0ea855773f5d8277318a78733b632f4b51bfcbf authored about 1 year ago by pancake <[email protected]>
Fix systemprompt varname

24d76dfef7afdeefe2e51a03138db493f72d314a authored about 1 year ago by pancake <[email protected]>
Make code compression algorithm recursive

b1ba080e395aa84504471b7b96b8021035c404ce authored about 1 year ago by pancake <[email protected]>
Add function call example

0908d07072072c7f559b79c594b6b1435608b49d authored about 1 year ago by pancake <[email protected]>
Support parsing large code (> 4KB)

74e9d15997ae8d1b2ab051b2e478519e2a55f478 authored about 1 year ago by pancake <[email protected]>
Fix prompt and internal chat log for llama models

8f9b7a89c8345990d3e3230612b7752a3b4eee7a authored about 1 year ago by pancake <[email protected]>
Also expose the self.model (WIP)

f5354006f9c4dabf0718d8e606da6ca454d6e955 authored about 1 year ago by pancake <[email protected]>
Protect env and report better errors when using unexistent ones

95bf393d8d3018c5ef7216ba35061cbabd31bd4f authored about 1 year ago by pancake <[email protected]>
Make context window, max tokens and temperature configurable

2cf08f4949d80c7112c413a06a451e58494c1a05 authored about 1 year ago by pancake <[email protected]>
Add support for scripting

1f850701da3d20ce8f921313ed22ce52a3c3cd5d authored about 1 year ago by pancake <[email protected]>
Initial PoC for the `aikeywords` implementation

a6f8cabdd1769c41a5c5158f1df212d23dc7d6a4 authored about 1 year ago by pancake <[email protected]>
vectordb fails if nothing is indexed :{

a50b163373da87b94ffce445baf281161398e647 authored about 1 year ago by pancake <[email protected]>
Makefile fails if not r2 is available

1fee6b29ae8322640071f0095a5f4aa2713c4193 authored about 1 year ago by pancake <[email protected]>
Fix the role prompt for alpaca models

efb473a884cc2ea2b8f5ae01b1424b8b56ddedb0 authored about 1 year ago by pancake <[email protected]>
Deprecate self.withresponse and just use env.chat.reply

399fccc0dd7ad5a9d68f21a2766b8838260a3dd2 authored about 1 year ago by pancake <[email protected]>
Fix vectordb usage when mprt is not available (macOS)

5b88cbae9db4784917abbdeec9ead11344e8e6ac authored about 1 year ago by pancake <[email protected]>
Use the best indexer model, be verbose and honor MAXMATCHES

20944643854af88b568de871f54c2a4428e6079b authored about 1 year ago by pancake <[email protected]>
Support indexing subdirectories and malformed markdown files

22ddf1e2784aef611c28caab335877f21e593b5d authored about 1 year ago by pancake <[email protected]>
Add hint support for llama models and fix the vectordb integration

0ea7ac905e6eb729841020f5039a19b045e18455 authored about 1 year ago by pancake <[email protected]>
Fix chat.trim usage bug

b1fc0c4ec2f4d20fbbaf03cb6890c7e62cc1333e authored about 1 year ago by pancake <[email protected]>
Add support for VectorDB (-e data.vectordb=true)

2a34bc145f43c8a0182a50b16856b8817d38b652 authored about 1 year ago by pancake <[email protected]>
Add buggy esil2c role

e24fdcaa30caec69707ac6b97e9b65e83943d62e authored about 1 year ago by pancake <[email protected]>
Move system message file into the doc/role

feea04b6124c15e15329e8520f1f36489e94ff64 authored about 1 year ago by pancake <[email protected]>
Debug mode doesnt imply llama logging

8f44e5fa45834903254ffcdcb53749bdb566a13e authored about 1 year ago by pancake <[email protected]>
Fix index.py when imported outside the project

3a717766fae0d302c3ce463cd33d451c2dea1b84 authored about 1 year ago by pancake <[email protected]>
Initial support for festival tts

27edd5be4f1c906ed80f29b2955fab42d417a346 authored about 1 year ago by pancake <[email protected]>
Load openai key from home

da447286907ffe2e4cab1dcbf72472b2716695ac authored about 1 year ago by pancake <[email protected]>
Helper to filter messages when prompt is too large

7fee94af8c266778b366dd28040edb21691a7c2b authored about 1 year ago by pancake <[email protected]>
Introduce chat.reply and dont include llama when using openai

c07f5a0a5ca80084489ff6df03d1a2ab2dc48617 authored about 1 year ago by pancake <[email protected]>
Improve code snippet with -c and -i

ec936edb9b2bd933c6e8a1f480b16ba6c73354c8 authored about 1 year ago by pancake <[email protected]>
Remove tokentrim dependency

6872e3a6c0c89ace1aab5987a7a14684a59da84d authored about 1 year ago by pancake <[email protected]>
Add data.local boolean config var to obey data.path or not

ead2bce9dd5596de3ef6f23c111b1490213dbf29 authored about 1 year ago by pancake <[email protected]>
improve mistral support

d880bd730a9f2accafd0e1b1de0cd3cdbf6b2c27 authored about 1 year ago by pancake <[email protected]>
Avoid dupplicated entries in the indexer

75ac1f2ddfce29a140ea8dce2d6447f36a0c605f authored about 1 year ago by pancake <[email protected]>
-e cannot delete keys anymore

b63832cf0ee449db91dcc8c640ba66e834d9e618 authored about 1 year ago by pancake <[email protected]>
Fix exception spotted when using the Mistral

caeafe42c8fe1aea62b598062fd6f939a8ce7284 authored about 1 year ago by pancake <[email protected]>
Release 0.4.0

dca854333556ffd10435a4ec9d36f4ee6943fcb5 authored about 1 year ago by pancake <[email protected]>
Fix startup when no r2pipe or r2lang installed

8ce41517fdb61b82879374838816f718c074ec2c authored about 1 year ago by pancake <[email protected]>
Add unidecode dependency

7ae32beb37b65bc2401f7c6dca153d078871fc11 authored about 1 year ago by pancake <[email protected]>
Fix requirements

dcc36345d0d931dbc1e08284bbd2e3992cb1ab4f authored about 1 year ago by pancake <[email protected]>
Honor voice.lang to support en,es,ca

17f56dfef6d877ffa53bdc63ad14614ffe697098 authored about 1 year ago by pancake <[email protected]>
Better queries with indexed data

7e3611a4ab889043e3fa441587a73031bc99f3a2 authored about 1 year ago by pancake <[email protected]>
Add more debugging logs when -e debug=true

dce96b3c6e91a4e2bcf5eeeee25443f34a9e04fc authored about 1 year ago by pancake <[email protected]>
Add hint type messages which speedup and improve mastodon requests

847d4fc5a5f746803cbbe312a9f7e0d3a46c95a7 authored about 1 year ago by pancake <[email protected]>
Move chat.bubble and chat.live into eval vars

becbf8bbd96fd864dc77eb20291b8281df33b94a authored about 1 year ago by pancake <[email protected]>
Honor chat.bubble eval var

40bd840ed7186d4d4b679f8727dcf7c8e74ac4c1 authored about 1 year ago by pancake <[email protected]>