nvicode 0.1.10 → 0.1.11

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.md +9 -87
  2. package/dist/cli.js +0 -0
  3. package/package.json +2 -2
package/README.md CHANGED
@@ -1,108 +1,30 @@
1
- # NviCode - Introducing one click Nvidia/OpenRouter keys to Claude Code. Free Claude code.
1
+ # NviCode
2
2
 
3
- Run Claude Code through NVIDIA-hosted models or OpenRouter using a simple CLI wrapper.
4
- Use top open-source model APIs on NVIDIA Build for free, with `nvicode` paced to `40 RPM` by default.
3
+ Connect Claude Code to NVIDIA or OpenRouter.
5
4
 
6
- Supported environments:
7
- - macOS
8
- - Ubuntu/Linux
9
- - WSL
10
- - Native Windows with Claude Code installed and working from PowerShell, CMD, or Git Bash
11
-
12
- ## Quickstart
13
-
14
- Install the published package:
5
+ ## Install
15
6
 
16
7
  ```sh
17
8
  npm install -g nvicode
18
9
  ```
19
10
 
20
- Set up provider, key, and model:
11
+ ## Setup
21
12
 
22
13
  ```sh
23
14
  nvicode select model
24
15
  ```
25
16
 
26
- The setup flow asks for provider, API key, and model:
17
+ NVIDIA keys: [https://build.nvidia.com/settings/api-keys](https://build.nvidia.com/settings/api-keys)
27
18
 
28
- - NVIDIA: get a free key from [NVIDIA Build API Keys](https://build.nvidia.com/settings/api-keys)
29
- - OpenRouter: use your OpenRouter API key
30
-
31
- Launch Claude Code through your selected provider:
19
+ ## Launch
32
20
 
33
21
  ```sh
34
22
  nvicode launch claude
35
23
  ```
36
24
 
37
- The first successful `nvicode launch claude` also installs persistent plain-`claude` routing.
38
- After that, restarting your terminal or PC and running:
39
-
40
- ```sh
41
- claude
42
- ```
43
-
44
- will continue using your selected `nvicode` provider and model.
45
-
46
- ## Screenshots
47
-
48
- ### Save your API key
49
-
50
- ![nvicode auth](https://raw.githubusercontent.com/dineshpotla/nvicode/main/assets/screenshots/auth.png)
51
-
52
- ### Choose a model
53
-
54
- ![nvicode select model](https://raw.githubusercontent.com/dineshpotla/nvicode/main/assets/screenshots/select-model.png)
55
-
56
- ### Launch Claude Code through your selected provider
57
-
58
- ![nvicode launch claude](https://raw.githubusercontent.com/dineshpotla/nvicode/main/assets/screenshots/launch.png)
59
-
60
- ## Commands
61
-
62
- Useful commands:
63
-
64
- ```sh
65
- nvicode dashboard
66
- nvicode usage
67
- nvicode activity
68
- nvicode models
69
- nvicode config
70
- nvicode auth
71
- nvicode launch claude -p "Reply with exactly OK"
72
- ```
73
-
74
- Provider behavior:
75
- - NVIDIA: starts a local proxy on `127.0.0.1:8788`, points Claude Code at it with `ANTHROPIC_BASE_URL`, and forwards requests to NVIDIA `chat/completions`.
76
- - OpenRouter: points Claude Code directly at `https://openrouter.ai/api` using OpenRouter credentials and Anthropic-compatible model ids.
77
-
78
- In an interactive terminal, `nvicode usage` refreshes live every 2 seconds. When piped or redirected, it prints a single snapshot.
79
-
80
- `nvicode select model` now asks for provider, optional API key update, and model choice in one guided flow.
81
- If no API key is saved for the active provider yet, `nvicode` prompts for one on first use.
82
- By default, the proxy paces upstream NVIDIA requests at `40 RPM`. Override that with `NVICODE_MAX_RPM` if your account has a different limit.
83
- The usage dashboard compares your local NVIDIA run cost against Claude Opus 4.6 at `$5 / MTok input` and `$25 / MTok output`, based on Anthropic pricing as of `2026-03-30`.
84
- If your NVIDIA endpoint is not free, override local cost estimates with `NVICODE_INPUT_USD_PER_MTOK` and `NVICODE_OUTPUT_USD_PER_MTOK`.
85
- Local `usage`, `activity`, and `dashboard` commands are available for NVIDIA proxy sessions. OpenRouter sessions use OpenRouter's direct connection path instead.
25
+ After the first successful launch, plain `claude` continues to use the selected `nvicode` provider and model.
86
26
 
87
27
  ## Requirements
88
28
 
89
- - Claude Code must already be installed on the machine.
90
- - Node.js 20 or newer is required to install `nvicode`.
91
- - On native Windows, Claude Code itself requires Git for Windows. See the [Claude Code setup docs](https://code.claude.com/docs/en/setup).
92
-
93
- ## Local Development
94
-
95
- These steps are only for contributors working from a git checkout. End users do not need them.
96
-
97
- ```sh
98
- npm install
99
- npm run build
100
- npm link
101
- ```
102
-
103
- ## Notes
104
-
105
- - `thinking` is disabled by default because some NVIDIA reasoning models can consume the entire output budget and return no visible answer to Claude Code.
106
- - The proxy supports basic text, tool calls, tool results, and token count estimation.
107
- - The proxy includes upstream request pacing and retries on NVIDIA `429` responses.
108
- - Claude Code remains the frontend; the selected provider/model becomes the backend.
29
+ - Claude Code must already be installed
30
+ - Node.js 20 or newer
package/dist/cli.js CHANGED
File without changes
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "nvicode",
3
- "version": "0.1.10",
3
+ "version": "0.1.11",
4
4
  "description": "NviCode - Introducing one click Nvidia/OpenRouter keys to Claude Code. Free Claude code.",
5
5
  "author": "Dinesh Potla",
6
6
  "keywords": [
@@ -17,7 +17,7 @@
17
17
  "nvicode": "dist/cli.js"
18
18
  },
19
19
  "scripts": {
20
- "build": "tsc -p tsconfig.json",
20
+ "build": "node -e \"require('fs').rmSync('dist',{recursive:true,force:true,maxRetries:10,retryDelay:100})\" && tsc -p tsconfig.json",
21
21
  "prepack": "npm run build",
22
22
  "postinstall": "node dist/postinstall.js",
23
23
  "typecheck": "tsc --noEmit",