Commits

Breadcrumbs

History for

onmain

User selector

All users

All time

Commit History

Commits on Apr 23, 2026

• Fix: Exclude chat_template for Mistral tokenizers in NVIDIA NIM (#130) (#131)

  Show description for 4fdf7e8
  

  anujbharambe

  and

  Alishahryar1

  authored
  4fdf7e8

  Copy full SHA for 4fdf7e8

• bug: nvidia didn't not support reasoning_budget parameter (#126)

  Show description for 4afca05
  

  3 peopleauthored
  4afca05

  Copy full SHA for 4afca05

• feat: add proxy support for httpx clients (#125)

  Show description for 2fe15bd
  

  arssing

  and

  Alishahryar1

  authored
  2fe15bd

  Copy full SHA for 2fe15bd

• feat: deepseek api support (#118)

  Show description for e719e4a
  

  duckinzzz

  and

  Alishahryar1

  authored
  e719e4a

  Copy full SHA for e719e4a

Commits on Apr 18, 2026

• Fixes for issue 113 and 116

  Alishahryar1

  committed
  835d045

  Copy full SHA for 835d045

Commits on Mar 28, 2026

• Gate NIM thinking params behind NIM_ENABLE_THINKING env var

  Show description for b75f47b
  Alishahryar1

  committed
  b75f47b

  Copy full SHA for b75f47b

Commits on Mar 26, 2026

• Support both kimi (thinking) and nemotron (enable_thinking) in chat_template_kwargs

  Alishahryar1

  committed
  2fad4dd

  Copy full SHA for 2fad4dd

• Fix NVIDIA NIM reasoning params for updated API

  Show description for f9e7f65
  Alishahryar1

  committed
  f9e7f65

  Copy full SHA for f9e7f65

Commits on Mar 23, 2026

• fix: remove unsupported include_stop_str_in_output NIM param (#95)

  yetval

  authored
  0003820

  Copy full SHA for 0003820

Commits on Mar 11, 2026

• removed logging utils

  Alishahryar1

  committed
  55945df

  Copy full SHA for 55945df

Commits on Mar 8, 2026

• feat: add llama.cpp provider for local anthropic messages API

  Alishahryar1

  committed
  5a36a32

  Copy full SHA for 5a36a32

• fix(providers): map httpx exceptions natively and remove type ignores

  Alishahryar1

  committed
  1aedf47

  Copy full SHA for 1aedf47

• feat(lmstudio): route natively to Anthropic /v1/messages endpoint

  Show description for 87d8ce1
  Alishahryar1

  committed
  87d8ce1

  Copy full SHA for 87d8ce1

• Move nim_settings from shared base class to NvidiaNimProvider (#78)

  Alishahryar1

  authored
  f57598f

  Copy full SHA for f57598f

Commits on Mar 1, 2026

• Improve deterministic error surfacing across stream and API

  Alishahryar1

  committed
  3475751

  Copy full SHA for 3475751

• Add code review fix plan covering 11 issues across modularity, encapsulation, performance, and dead code (#62)

  Alishahryar1

  authored
  aee9f0a

  Copy full SHA for aee9f0a

Commits on Feb 28, 2026

• Major cleanup with GLM-5

  Alishahryar1

  committed
  744eec2

  Copy full SHA for 744eec2

• Major refactor done with minimax m2.5

  Alishahryar1

  committed
  a74ec74

  Copy full SHA for a74ec74

• minor refactor using minimax m2.5

  Alishahryar1

  committed
  79a1ae0

  Copy full SHA for 79a1ae0

• Backup/before cleanup 20260222 230402 (#58)

  Alishahryar1

  authored
  c4d8681

  Copy full SHA for c4d8681

Commits on Feb 20, 2026

• moved text.py to common utils for providers

  Alishahryar1

  committed
  2b0495d

  Copy full SHA for 2b0495d

Commits on Feb 19, 2026

• Make PROVIDER_MAX_CONCURRENCY required with default of 5

  Show description for 45b7e4c
  claude

  committed
  45b7e4c

  Copy full SHA for 45b7e4c

• Add queue-level concurrency limit to provider streaming

  Show description for afaf50a
  claude

  committed
  afaf50a

  Copy full SHA for afaf50a

• Fixing sse stream

  suryawanshishantanu6

  committed
  24a5e4d

  Copy full SHA for 24a5e4d

Commits on Feb 18, 2026

• new linter rules and fixes

  Alishahryar1

  committed
  b05d0d2

  Copy full SHA for b05d0d2

Commits on Feb 17, 2026

• Phase 4-6: Dead code removal, performance, minor fixes

  Show description for bfc781e
  

  cursoragent

  and

  Alishahryar1

  committed
  bfc781e

  Copy full SHA for bfc781e

• Phase 3: Fix encapsulation violations

  Show description for 72b7e34
  

  cursoragent

  and

  Alishahryar1

  committed
  72b7e34

  Copy full SHA for 72b7e34

• Phase 2: Extract OpenAICompatibleProvider base class

  Show description for bab86a2
  

  cursoragent

  and

  Alishahryar1

  committed
  bab86a2

  Copy full SHA for bab86a2

• Phase 1: Extract shared provider utils into providers/common/

  Show description for d3f5f88
  

  cursoragent

  and

  Alishahryar1

  committed
  d3f5f88

  Copy full SHA for d3f5f88

Commits on Feb 16, 2026

• Add configurable HTTP timeouts for provider API requests

  Show description for 01852e1
  Alishahryar1

  committed
  01852e1

  Copy full SHA for 01852e1

• Updated thinking keys

  Alishahryar1

  committed
  1d52e5d

  Copy full SHA for 1d52e5d

• Refactor done using GLM-5

  Alishahryar1

  committed
  539854f

  Copy full SHA for 539854f

• Add LM Studio provider support

  Show description for b83be84
  Alishahryar1

  committed
  b83be84

  Copy full SHA for b83be84

Commits on Feb 15, 2026

• Updated open router default max tokens

  Alishahryar1

  committed
  30ea67b

  Copy full SHA for 30ea67b

• feat: add OpenRouter support and configuration options

  Show description for e5a0960
  Alishahryar1

  committed
  e5a0960

  Copy full SHA for e5a0960