Skip to content
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

Add Llama 3.3 #250

Merged
merged 2 commits into from
Dec 6, 2024
Merged

Add Llama 3.3 #250

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0588f99
Add Llama 3.3
svilupp Dec 6, 2024
e7b5392
update aliases
svilupp Dec 6, 2024
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
6 changes: 4 additions & 2 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,12 +10,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0

### Fixed

## [0.65.2]
## [0.66.0]

### Added
- Added support for Groq's new Llama3.3 models. Updated `gllama370`, `gl70`, `glm` aliases to `llama-3.3-70b-versatile` and added `gl70s`, `glms` aliases to `llama-3.3-70b-specdec` (faster with speculative decoding).

### Fixed
- Fixed a bug in `extract_docstring` where it would not correctly block "empty" docstrings on Julia 1.11.


## [0.65.1]

### Fixed
Expand Down
2 changes: 1 addition & 1 deletion Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
name = "PromptingTools"
uuid = "670122d1-24a8-4d70-bfce-740807c42192"
authors = ["J S @svilupp and contributors"]
version = "0.65.2"
version = "0.66.0"

[deps]
AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
Expand Down
18 changes: 15 additions & 3 deletions src/user_preferences.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -432,14 +432,16 @@ aliases = merge(
## Groq
"gllama3" => "llama-3.1-8b-instant",
"gl3" => "llama-3.1-8b-instant",
"gllama370" => "llama-3.1-70b-versatile",
"gl70" => "llama-3.1-70b-versatile",
"gllama370" => "llama-3.3-70b-versatile",
"gl70" => "llama-3.3-70b-versatile",
"gl70s" => "llama-3.3-70b-specdec", #s for small speculative decoding
"gllama3405" => "llama-3.1-405b-reasoning",
"gl405" => "llama-3.1-405b-reasoning",
"glxxs" => "llama-3.2-1b-preview", #xxs for extra extra small
"glxs" => "llama-3.2-3b-preview", #xs for extra small
"gls" => "llama-3.1-8b-instant", #s for small
"glm" => "llama-3.1-70b-versatile", #m for medium
"glm" => "llama-3.3-70b-versatile", #m for medium
"glms" => "llama-3.3-70b-specdec", #ms for medium speculative decoding
"gll" => "llama-3.1-405b-reasoning", #l for large
"gmixtral" => "mixtral-8x7b-32768",
"ggemma9" => "gemma2-9b-it",
Expand Down Expand Up @@ -943,6 +945,16 @@ registry = Dict{String, ModelSpec}(
2.4e-5,
"Anthropic's Claude 2.1 model."),
## Groq -- using preliminary pricing on https://wow.groq.com/
"llama-3.3-70b-specdec" => ModelSpec("llama-3.3-70b-specdec",
GroqOpenAISchema(),
5.9e-7,
9.9e-7,
"Meta's Llama3.3 70b with speculative decoding (slightly more expensive), hosted by Groq. Max output 8192 tokens, 128K context. See details [here](https://console.groq.com/docs/models)"),
"llama-3.3-70b-versatile" => ModelSpec("llama-3.3-70b-versatile",
GroqOpenAISchema(),
5.9e-7,
7.9e-7,
"Meta's Llama3.3 70b, hosted by Groq. Max output 8192 tokens, 128K context. See details [here](https://console.groq.com/docs/models)"),
"llama-3.1-405b-reasoning" => ModelSpec("llama-3.1-405b-reasoning",
GroqOpenAISchema(),
5e-6, # based on prices at together.ai... likely it will be much cheaper
Expand Down
Loading