Commit graph

  • 9a7d430ff2
    tests : disable obsolete Georgi Gerganov 2024-04-29 14:12:34 +03:00
  • 120cf37d54
    models : add phi-3, mpt, gpt-2, starcoder Georgi Gerganov 2024-04-29 13:40:30 +03:00
  • 603f18bc46 feat: small changes to allow jina embeddings ZH model Joan Martinez 2024-04-29 12:23:20 +02:00
  • a0c870db85 Fix merge jaime-m-p 2024-04-29 11:09:52 +02:00
  • 866e3941f7
    Merge branch 'ggerganov:gg/bpe-preprocess' into gg/bpe-preprocess jaime-m-p 2024-04-29 10:55:15 +02:00
  • c21ab1833e
    scripts : ignore new update script in check-requirements.sh Georgi Gerganov 2024-04-29 11:24:05 +03:00
  • af05268cdd
    unicode : cleanup Georgi Gerganov 2024-04-29 11:20:42 +03:00
  • c68d2596ea
    tests : add more vocabs and tests Georgi Gerganov 2024-04-29 11:07:25 +03:00
  • 43708d22c3
    tests : refactor vocab tests Georgi Gerganov 2024-04-29 10:46:43 +03:00
  • ef4cca9e87
    cmake : refactor test targets Georgi Gerganov 2024-04-29 09:53:14 +03:00
  • 098e7bbc32
    Update README-sycl.md Olivier Chafik 2024-04-29 07:53:30 +01:00
  • 03aa7f80e4 Add ability to analyze multiple images on a single command line without having the reload the model root 2024-04-29 03:03:59 +00:00
  • 49e078f79d changed array size parameters to size_t l3utterfly 2024-04-29 10:58:26 +09:00
  • c33bdf397d fix improper function signature Christian Zhou-Zheng 2024-04-28 21:52:33 -04:00
  • d701e81b15 bugfix: Omitted the colon, resulting in the failure of execution of convert-hf-to-gguf.py XingXing Qiao 2024-04-29 09:48:00 +08:00
  • 3caec6bb41 removed unused llama_context in dry sampler l3utterfly 2024-04-29 10:25:25 +09:00
  • 793e1e221b updated header def for dry sampler to match implementation l3utterfly 2024-04-29 10:22:58 +09:00
  • 85dadac483 added parameter for DRY penalty range, separate from the original repetition penalty range l3utterfly 2024-04-29 10:20:17 +09:00
  • 75beda2a84 fixed various issues with sampler pointed out by original creator l3utterfly 2024-04-29 10:01:50 +09:00
  • 663d274bbf ci : add building in MSYS2 environments (Windows) Przemyslaw Pawelczyk 2024-04-29 02:01:20 +02:00
  • 0cf9ed3457 Restore BOM jaime-m-p 2024-04-29 01:35:08 +02:00
  • c010ce5274 Set console width constant TevinWang 2024-04-28 18:52:01 -04:00
  • 6d5c7f0903 Add windows support for console width TevinWang 2024-04-28 18:49:09 -04:00
  • 9d6f198bfe
    Update llava-cli.cpp cpumaxx 2024-04-28 15:44:32 -07:00
  • e00b4a8f81
    Fix more int overflow during quant (PPL/CUDA). (#6563) b2755 DAN™ 2024-04-28 18:38:44 -04:00
  • 0258f9bd3d Revert back to int64_t. DAN™ 2024-04-28 18:33:37 -04:00
  • 5e906de275
    Update llava-cli.cpp cpumaxx 2024-04-28 15:32:32 -07:00
  • e441cc8992
    Update examples/llava/llava-cli.cpp cpumaxx 2024-04-28 15:29:45 -07:00
  • 702a744670 Move split functionality to new GGUFManager class Christian Azinn 2024-04-28 18:22:32 -04:00
  • 91c10ef225 Fix some more int overflow in softmax. DAN™ 2024-04-28 17:07:41 -04:00
  • 2a48873914 Typing jaime-m-p 2024-04-29 00:12:56 +02:00
  • 763460ba71 Use a global progress callback instead of completing removing progress TevinWang 2024-04-28 17:54:39 -04:00
  • 063e201b02 Fixes, Linux support over 64 CPUs, Core 0 enabled at 6 cores and below mann1x 2024-04-28 22:46:12 +02:00
  • 6e4d2af6c3 already exists unicode_tolower() jaime-m-p 2024-04-28 21:57:22 +02:00
  • 7b1210f6a8
    lint : fix Georgi Gerganov 2024-04-28 22:51:13 +03:00
  • 5cc4b2cf01 Using char32_t for codepoints jaime-m-p 2024-04-28 21:51:12 +02:00
  • 78081502e9
    convert : exercise contractions Georgi Gerganov 2024-04-28 22:18:20 +03:00
  • 0f9058ceec
    convert : add comments Georgi Gerganov 2024-04-28 22:10:04 +03:00
  • 02fd977fe1
    convert : remove unused functions Georgi Gerganov 2024-04-28 22:03:21 +03:00
  • e8dd4a1494
    lint : fix Georgi Gerganov 2024-04-28 22:02:10 +03:00
  • 491f2339bb
    lint : fix Georgi Gerganov 2024-04-28 21:42:58 +03:00
  • 1545550ec2
    unicode : normalize signatures Georgi Gerganov 2024-04-28 21:40:36 +03:00
  • 1c888eb4da
    convert : add falcon Georgi Gerganov 2024-04-28 21:26:40 +03:00
  • 4e3e6d8ecc
    lint : update Georgi Gerganov 2024-04-28 21:16:50 +03:00
  • 7642973616
    convert : add convert-hf-to-gguf-update.py Georgi Gerganov 2024-04-28 20:29:32 +03:00
  • e11fe2fb6a llama3 custom regex split jaime-m-p 2024-04-28 19:27:06 +02:00
  • f1a2eb8ec7 build: update MinGW cmake commands Olivier Chafik 2024-04-28 18:06:43 +01:00
  • ba7f8af90e build: remove another superfluous . Olivier Chafik 2024-04-28 18:03:27 +01:00
  • dc5d7fee9d build: drop implicit . from cmake config command Olivier Chafik 2024-04-28 17:59:57 +01:00
  • 3e8869697c
    build: fix typo Olivier Chafik 2024-04-28 17:57:35 +01:00
  • ee73eb7915 make : change GNU make default CXX from g++ to c++ Przemyslaw Pawelczyk 2024-04-28 18:51:51 +02:00
  • d39085440d readme: cmake . -B build && cmake --build build ochafik 2024-04-28 16:04:49 +01:00
  • ee6d1b3fb4
    unicode : simplify Georgi Gerganov 2024-04-28 18:36:57 +03:00
  • 7bb36ccf91
    gguf : enforce that tensor names are unique (#6905) b2754 Xuan Son Nguyen 2024-04-28 17:36:18 +02:00
  • e972e6cbf8
    unicode : clean-up Georgi Gerganov 2024-04-28 18:01:59 +03:00
  • ce023f6f2f
    add device version in device list (#6959) b2753 Neo Zhang 2024-04-28 22:40:31 +08:00
  • c70037f2b3 grammars: nits (revert const grammar sig, fix comment) ochafik 2024-04-28 15:34:56 +01:00
  • d3425f5cf1 grammar: nit (layout) ochafik 2024-04-28 15:28:58 +01:00
  • 49f0faaa0e grammars: cache codepoints in llama_new_context_with_model ochafik 2024-04-28 15:24:35 +01:00
  • d41f314740 grammars: move token caches to llama_context ochafik 2024-04-28 14:55:28 +01:00
  • d63cc9068b
    Merge branch 'master' into gg/bpe-preprocess Georgi Gerganov 2024-04-28 15:34:36 +03:00
  • 53e505d504 use std::random_device{}() for default random seed David Renshaw 2024-04-28 08:26:46 -04:00
  • b97add52a4
    unicode : category support via std::regex Georgi Gerganov 2024-04-28 13:42:00 +03:00
  • 6e472f58e4 flake.lock: Update github-actions[bot] 2024-04-28 00:18:27 +00:00
  • b1df2f212e
    throw exception inside llama_model_loader Xuan Son Nguyen 2024-04-28 11:16:34 +02:00
  • 1917730a2c add chatglm3-6b model support XingXing Qiao 2024-04-28 10:55:57 +08:00
  • e898fd9fda add device version in device list arthw 2024-04-28 16:02:44 +08:00
  • 0c115da251 fix typo Paulo de Castro 2024-04-28 00:37:53 -03:00
  • 91d94eeebd server: avoid breaking KV cache when prompt >= n_ctx Paulo de Castro 2024-04-28 00:34:07 -03:00
  • 33c60537e8
    Allow params.js to be embedded in server.cpp xatier 2024-04-24 20:35:57 -05:00
  • c724e32573
    server: public: move default parameters to its own file xatier 2024-04-23 20:21:52 -05:00
  • 72cbd4e014 Support split by size and dry run to write estimated shards/filesizes Christian Azinn 2024-04-27 21:14:41 -04:00
  • 6134134080 use std::chrono::system_clock::now() for random seed David Renshaw 2024-04-27 20:37:12 -04:00
  • 5faf92c438 flake.lock: Update github-actions[bot] 2024-04-28 00:18:27 +00:00
  • 9f13623149 Merge remote-tracking branch 'origin/master' into grammar-fast ochafik 2024-04-27 23:49:34 +01:00
  • b4a00cec0f Merge branch 'gguf-read' into agent-example ochafik 2024-04-27 23:17:27 +01:00
  • 8d503ef482 grammars: faster llama_grammar_copy ochafik 2024-04-21 15:52:25 +01:00
  • 00c709eb4a grammars: cache decoded tokens ochafik 2024-04-21 15:52:16 +01:00
  • 09c256594d grammars: early exit when no next_candidates to reject ochafik 2024-04-21 01:12:05 +01:00
  • 0120f7cc95 agent: fix wait --std-tools Olivier Chafik 2024-04-10 19:47:01 +01:00
  • 89dcc062a4 agent: mypy type fixes Olivier Chafik 2024-04-10 19:45:13 +01:00
  • ea0c31b10b agent: ensure DATA_DIR exists Olivier Chafik 2024-04-10 11:39:35 +01:00
  • a98f48315c agent: python tool: return errors ochafik 2024-04-10 09:27:25 +01:00
  • f9afb041e2 agent: python tool: test serializability of variables ochafik 2024-04-10 09:14:24 +01:00
  • 082d54db14 agent: rename fake weather tools ochafik 2024-04-10 08:37:09 +01:00
  • 6c00378630 agent: nits ochafik 2024-04-10 01:08:07 +01:00
  • 1475b1eefa agent: fix killing of subprocesses ochafik 2024-04-10 08:05:03 +01:00
  • 24e34f174b agent: nit ochafik 2024-04-10 01:29:20 +01:00
  • a61ebebaa0 agent: hint at math import in python tool ochafik 2024-04-10 01:22:09 +01:00
  • 9fe269e24a openai: nit ochafik 2024-04-10 01:08:07 +01:00
  • a634e03aba agent: cache_prompt=True ochafik 2024-04-10 01:03:39 +01:00
  • 0532680f40 agent: nits Olivier Chafik 2024-04-09 23:50:08 +01:00
  • 6880f1d4c0 agent: support basic openapi tools (incl. from fastify sandbox) Olivier Chafik 2024-04-09 23:40:11 +01:00
  • 85820f4401 agent: fix sandbox dockerfile Olivier Chafik 2024-04-09 21:03:32 +01:00
  • b447a743fb agent: revert to json schemas (ts not ready for refs) ochafik 2024-04-09 09:41:45 +01:00
  • 701a66d80f agent: fix response_format ochafik 2024-04-09 02:14:08 +01:00
  • 6e52a9ce48 Update test_chat_handlers.md ochafik 2024-04-08 19:18:01 +01:00
  • 22fe86d8b8 openai tools: TS signatures work well too at a fraction of the eval cost ochafik 2024-03-30 02:57:36 +00:00
  • 19811a4011 openai: tests didn't catch output format ochafik 2024-03-30 02:24:04 +00:00
  • 09de4eb9ed openai: actually use thoughtful examples in tests ochafik 2024-03-30 01:57:05 +00:00