Skip to content

fix gpt-5, and newer openai models that require maxcompletiontokens. #2264

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

Merged
sawka merged 1 commit into from
Aug 19, 2025
Merged

fix gpt-5, and newer openai models that require maxcompletiontokens. #2264

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
55 changes: 49 additions & 6 deletions docs/docs/ai-presets.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,6 +70,22 @@ To use Claude models, create a preset like this:
}
```

### OpenAI

To use OpenAI's models:

```json
{
"ai@openai-gpt41": {
"display:name": "GPT-4.1",
"display:order": 2,
"ai:*": true,
"ai:model": "gpt-4.1",
"ai:apitoken": "<your OpenAI API key>"
}
}
```

### Local LLMs (Ollama)

To connect to a local Ollama instance:
Expand All @@ -78,7 +94,7 @@ To connect to a local Ollama instance:
{
"ai@ollama-llama": {
"display:name": "Ollama - Llama2",
"display:order": 2,
"display:order": 3,
"ai:*": true,
"ai:baseurl": "http://localhost:11434/v1",
"ai:name": "llama2",
Expand All @@ -98,7 +114,7 @@ To connect to Azure AI services:
{
"ai@azure-gpt4": {
"display:name": "Azure GPT-4",
"display:order": 3,
"display:order": 4,
"ai:*": true,
"ai:apitype": "azure",
"ai:baseurl": "<your Azure AI base URL>",
Expand All @@ -118,7 +134,7 @@ To use Perplexity's models:
{
"ai@perplexity-sonar": {
"display:name": "Perplexity Sonar",
"display:order": 4,
"display:order": 5,
"ai:*": true,
"ai:apitype": "perplexity",
"ai:model": "llama-3.1-sonar-small-128k-online",
Expand All @@ -135,7 +151,7 @@ To use Google's Gemini models from [Google AI Studio](https://aistudio.google.co
{
"ai@gemini-2.0": {
"display:name": "Gemini 2.0",
"display:order": 5,
"display:order": 6,
"ai:*": true,
"ai:apitype": "google",
"ai:model": "gemini-2.0-flash-exp",
Expand All @@ -158,9 +174,16 @@ You can define multiple presets in your `ai.json` file:
"ai:model": "claude-3-5-sonnet-latest",
"ai:apitoken": "<your anthropic API key>"
},
"ai@openai-gpt41": {
"display:name": "GPT-4.1",
"display:order": 2,
"ai:*": true,
"ai:model": "gpt-4.1",
"ai:apitoken": "<your OpenAI API key>"
},
"ai@ollama-llama": {
"display:name": "Ollama - Llama2",
"display:order": 2,
"display:order": 3,
"ai:*": true,
"ai:baseurl": "http://localhost:11434/v1",
"ai:name": "llama2",
Expand All @@ -169,7 +192,7 @@ You can define multiple presets in your `ai.json` file:
},
"ai@perplexity-sonar": {
"display:name": "Perplexity Sonar",
"display:order": 3,
"display:order": 4,
"ai:*": true,
"ai:apitype": "perplexity",
"ai:model": "llama-3.1-sonar-small-128k-online",
Expand All @@ -187,3 +210,23 @@ Remember to set your default preset in `settings.json`:
"ai:preset": "ai@claude-sonnet"
}
```

## Using a Proxy

If you need to route AI requests through an HTTP proxy, you can add the `ai:proxyurl` setting to any preset:

```json
{
"ai@claude-with-proxy": {
"display:name": "Claude 3 Sonnet (via Proxy)",
"display:order": 1,
"ai:*": true,
"ai:apitype": "anthropic",
"ai:model": "claude-3-5-sonnet-latest",
"ai:apitoken": "<your anthropic API key>",
"ai:proxyurl": "http://proxy.example.com:8080"
}
}
```

The proxy URL should be in the format `http://host:port` or `https://host:port`. This setting works with all AI providers except Wave Cloud AI (the default).
5 changes: 3 additions & 2 deletions docs/docs/config.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,9 @@ wsh editconfig
| ai:model | string | model name to pass to API |
| ai:apiversion | string | for Azure AI only (when apitype is "azure", this will default to "2023-05-15") |
| ai:orgid | string | |
| ai:maxtokens | int | max tokens to pass to API |
| ai:timeoutms | int | timeout (in milliseconds) for AI calls |
| ai:maxtokens | int | max tokens to pass to API |
| ai:timeoutms | int | timeout (in milliseconds) for AI calls |
| ai:proxyurl | string | HTTP proxy URL for AI API requests (does not apply to Wave Cloud AI) |
| conn:askbeforewshinstall | bool | set to false to disable popup asking if you want to install wsh extensions on new machines |
| term:fontsize | float | the fontsize for the terminal block |
| term:fontfamily | string | font family to use for terminal block |
Expand Down
40 changes: 12 additions & 28 deletions pkg/waveai/openaibackend.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,14 @@ func defaultAzureMapperFn(model string) string {
return regexp.MustCompile(`[.:]`).ReplaceAllString(model, "")
}

func isReasoningModel(model string) bool {
m := strings.ToLower(model)
return strings.HasPrefix(m, "o1") ||
strings.HasPrefix(m, "o3") ||
strings.HasPrefix(m, "o4") ||
strings.HasPrefix(m, "gpt-5")
}

func setApiType(opts *wshrpc.WaveAIOptsType, clientConfig *openaiapi.ClientConfig) error {
ourApiType := strings.ToLower(opts.APIType)
if ourApiType == "" || ourApiType == APIType_OpenAI || ourApiType == strings.ToLower(string(openaiapi.APITypeOpenAI)) {
Expand Down Expand Up @@ -123,38 +131,14 @@ func (OpenAIBackend) StreamCompletion(ctx context.Context, request wshrpc.WaveAI
Messages: convertPrompt(request.Prompt),
}

// Handle o1 models differently - use non-streaming API
if strings.HasPrefix(request.Opts.Model, "o1-") {
// Set MaxCompletionTokens for reasoning models, MaxTokens for others
if isReasoningModel(request.Opts.Model) {
req.MaxCompletionTokens = request.Opts.MaxTokens
req.Stream = false

// Make non-streaming API call
resp, err := client.CreateChatCompletion(ctx, req)
if err != nil {
rtn <- makeAIError(fmt.Errorf("error calling openai API: %v", err))
return
}

// Send header packet
headerPk := MakeWaveAIPacket()
headerPk.Model = resp.Model
headerPk.Created = resp.Created
rtn <- wshrpc.RespOrErrorUnion[wshrpc.WaveAIPacketType]{Response: *headerPk}

// Send content packet(s)
for i, choice := range resp.Choices {
pk := MakeWaveAIPacket()
pk.Index = i
pk.Text = choice.Message.Content
pk.FinishReason = string(choice.FinishReason)
rtn <- wshrpc.RespOrErrorUnion[wshrpc.WaveAIPacketType]{Response: *pk}
}
return
} else {
req.MaxTokens = request.Opts.MaxTokens
}

// Original streaming implementation for non-o1 models
req.Stream = true
req.MaxTokens = request.Opts.MaxTokens
if request.Opts.MaxChoices > 1 {
req.N = request.Opts.MaxChoices
}
Expand Down
Loading