arcee-ai fastmlx issues

arcee-ai / fastmlx

FastMLX is a high performance production ready API to host MLX models.

Other

222 stars 25 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Tool parsing and tool choice

#40 cmcmaster1 opened 5 days ago
4
Add Qwen function calling

#39 cmcmaster1 opened 6 days ago
2
Fixed default prompt template in config.json

#38 cmcmaster1 closed 6 days ago
0
Fix Docs Link

#37 Blaizzy opened 3 weeks ago
0
Fix list models tests

#36 Blaizzy opened 3 weeks ago
0
OpenWebUI doesn't connect to FastMLX

#35 SwagMuffinMcYoloPants opened 1 month ago
5
/v1/models - format output to match OpenAI styling

#34 aricshow closed 3 weeks ago
4
ReadMe Broken Docs Link

#33 larson-carter closed 4 weeks ago
3
feat: open webui compliance

#32 viljark closed 2 months ago
3
Future: add image generation prompts?

#31 stewartugelow opened 3 months ago
1
Using OpenAI API compliant to support vision models

#30 madroidmaq closed 2 months ago
5
Multiple workers do not share memory, which causes a full model reload for each message generation.

#29 ZachZimm opened 3 months ago
5
Integrate mlx-hub like functionality?

#28 stewartugelow opened 3 months ago
2
Implement Token Usage Tracking

#27 ZachZimm closed 3 weeks ago
5
Memory leak ?

#26 iLoveBug opened 3 months ago
7
Fix tools template loader (Jinja 2 Template not found)

#25 Blaizzy closed 3 months ago
0
Implement CLI Client for FastMLX

#24 Blaizzy opened 3 months ago
0
FastMLX Python Client

#23 Blaizzy opened 3 months ago
3
Implement role:system in messages

#22 pablo-mano closed 3 months ago
3
Add support for tool calling

#21 Blaizzy closed 3 months ago
0
How to make it verbose?

#20 namp opened 4 months ago
3
Add Docs

#19 Blaizzy closed 4 weeks ago
2
Feature Request: Integrate Features from Ollama

#18 evertjr opened 4 months ago
8
Explore integration with exo?

#17 stewartugelow opened 4 months ago
4
Documentation link is 404

#16 awni closed 4 weeks ago
1
feat: Set `workers` through env variable, improved defaults

#15 SiddhantSadangi closed 4 months ago
2
Potential error in shutdown if manually cancelled

#14 stewartugelow closed 4 months ago
6
Microsoft Phi 3 EOS token not recognized

#13 stewartugelow closed 4 months ago
2
(0.1) Uvicorn running on http://0.0.0.0:8000

#12 stewartugelow closed 4 months ago
2
Weird image URL bug with Wikimedia

#11 stewartugelow closed 4 months ago
9
Implement Error Handling for Unsupported Model Types

#10 Blaizzy opened 4 months ago
0
Implement Model Loading State Tracker

#9 Blaizzy opened 4 months ago
0
Implement Basic Token Usage Tracking

#8 Blaizzy opened 4 months ago
1
Add Parallel calls usage

#7 Blaizzy closed 4 months ago
0
No chat template specified for llava models error

#6 stewartugelow opened 4 months ago
13
max_tokens not overriding the default

#5 stewartugelow closed 4 months ago
1
Add support for token streaming, parallel jobs and custom CORS

#4 Blaizzy closed 4 months ago
0
Documention link is broken

#3 ipfans closed 4 weeks ago
1
Cross origin support

#2 digicali closed 4 months ago
4
Setup FastMLX

#1 Blaizzy closed 4 months ago
0