mirror of
https://github.com/ollama/ollama.git
synced 2026-04-22 08:45:53 +02:00
* don't require pulling stubs for cloud models This is a first in a series of PRs that will better integrate Ollama's cloud into the API and CLI. Previously we used to have a layer of indirection where you'd first have to pull a "stub" model that contains a reference to a cloud model. With this change, you don't have to pull first, you can just use a cloud model in various routes like `/api/chat` and `/api/show`. This change respects <https://github.com/ollama/ollama/pull/14221>, so if cloud is disabled, these models won't be accessible. There's also a new, simpler pass-through proxy that doesn't convert the requests ahead of hitting the cloud models, which they themselves already support various formats (e.g., `v1/chat/completions` or Open Responses, etc.). This will help prevent issues caused by double converting (e.g., `v1/chat/completions` converted to `api/chat` on the client, then calling cloud and converting back to a `v1/chat/completions` response instead of the cloud model handling the original `v1/chat/completions` request first). There's now a notion of "source tags", which can be mixed with existing tags. So instead of having different formats like`gpt-oss:20b-cloud` vs. `kimi-k2.5:cloud` (`-cloud` suffix vs. `:cloud`), you can now specify cloud by simply appending `:cloud`. This PR doesn't change model resolution yet, but sets us up to allow for things like omitting the non-source tag, which would make something like `ollama run gpt-oss:cloud` work the same way that `ollama run gpt-oss` already works today. More detailed changes: - Added a shared model selector parser in `types/modelselector`: - supports `:cloud` and `:local` - accepts source tags in any position - supports legacy `:<tag>-cloud` - rejects conflicting source tags - Integrated selector handling across server inference/show routes: - `GenerateHandler`, `ChatHandler`, `EmbedHandler`, `EmbeddingsHandler`, `ShowHandler` - Added explicit-cloud passthrough proxy for ollama.com: - same-endpoint forwarding for `/api/*`, `/v1/*`, and `/v1/messages` - normalizes `model` (and `name` for `/api/show`) before forwarding - forwards request headers except hop-by-hop/proxy-managed headers - uses bounded response-header timeout - handles auth failures in a friendly way - Preserved cloud-disable behavior (`OLLAMA_NO_CLOUD`) - Updated create flow to support `FROM ...:cloud` model sources (though this flow uses the legacy proxy still, supporting Modelfile overrides is more complicated with the direct proxy approach) - Updated CLI/TUI/config cloud detection to use shared selector logic - Updated CLI preflight behavior so explicit cloud requests do not auto-pull local stubs What's next? - Cloud discovery/listing and cache-backed `ollama ls` / `/api/tags` - Modelfile overlay support for virtual cloud models on OpenAI/Anthropic request families - Recommender/default-selection behavior for ambiguous model families - Fully remove the legacy flow Fixes: https://github.com/ollama/ollama/issues/13801 * consolidate pull logic into confirmAndPull helper pullIfNeeded and ShowOrPull shared identical confirm-and-pull logic. Extract confirmAndPull to eliminate the duplication. * skip local existence checks for cloud models ModelExists and the TUI's modelExists both check the local model list, which causes cloud models to appear missing. Return true early for explicit cloud models so the TUI displays them beside the integration name and skips re-prompting the model picker on relaunch. * support optionally pulling stubs for newly-style names We now normalize names like `<family>:<size>:cloud` into legacy-style names like `<family>:<size>-cloud` for pulling and deleting (this also supports stripping `:local`). Support for pulling cloud models is temporary, once we integrate properly into `/api/tags` we won't need this anymore. * Fix server alias syncing * Update cmd/cmd.go Co-authored-by: Parth Sareen <parth.sareen@ollama.com> * address comments * improve some naming --------- Co-authored-by: ParthSareen <parth.sareen@ollama.com>
1083 lines
31 KiB
Go
1083 lines
31 KiB
Go
package server
|
|
|
|
import (
|
|
"bufio"
|
|
"bytes"
|
|
"context"
|
|
"encoding/json"
|
|
"errors"
|
|
"io"
|
|
"net/http"
|
|
"net/http/httptest"
|
|
"strings"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/gin-gonic/gin"
|
|
"github.com/ollama/ollama/api"
|
|
internalcloud "github.com/ollama/ollama/internal/cloud"
|
|
"github.com/ollama/ollama/middleware"
|
|
)
|
|
|
|
func TestStatusHandler(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
t.Setenv("OLLAMA_NO_CLOUD", "1")
|
|
|
|
s := Server{}
|
|
w := createRequest(t, s.StatusHandler, nil)
|
|
if w.Code != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d", w.Code)
|
|
}
|
|
|
|
var resp api.StatusResponse
|
|
if err := json.NewDecoder(w.Body).Decode(&resp); err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
|
|
if !resp.Cloud.Disabled {
|
|
t.Fatalf("expected cloud.disabled true, got false")
|
|
}
|
|
if resp.Cloud.Source != "env" {
|
|
t.Fatalf("expected cloud.source env, got %q", resp.Cloud.Source)
|
|
}
|
|
}
|
|
|
|
func TestCloudDisabledBlocksRemoteOperations(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
t.Setenv("OLLAMA_NO_CLOUD", "1")
|
|
|
|
s := Server{}
|
|
|
|
w := createRequest(t, s.CreateHandler, api.CreateRequest{
|
|
Model: "test-cloud",
|
|
RemoteHost: "example.com",
|
|
From: "test",
|
|
Info: map[string]any{
|
|
"capabilities": []string{"completion"},
|
|
},
|
|
Stream: &stream,
|
|
})
|
|
if w.Code != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d", w.Code)
|
|
}
|
|
|
|
t.Run("chat remote blocked", func(t *testing.T) {
|
|
w := createRequest(t, s.ChatHandler, api.ChatRequest{
|
|
Model: "test-cloud",
|
|
Messages: []api.Message{{Role: "user", Content: "hi"}},
|
|
})
|
|
if w.Code != http.StatusForbidden {
|
|
t.Fatalf("expected status 403, got %d", w.Code)
|
|
}
|
|
if got := w.Body.String(); got != `{"error":"`+internalcloud.DisabledError(cloudErrRemoteInferenceUnavailable)+`"}` {
|
|
t.Fatalf("unexpected response: %s", got)
|
|
}
|
|
})
|
|
|
|
t.Run("generate remote blocked", func(t *testing.T) {
|
|
w := createRequest(t, s.GenerateHandler, api.GenerateRequest{
|
|
Model: "test-cloud",
|
|
Prompt: "hi",
|
|
})
|
|
if w.Code != http.StatusForbidden {
|
|
t.Fatalf("expected status 403, got %d", w.Code)
|
|
}
|
|
if got := w.Body.String(); got != `{"error":"`+internalcloud.DisabledError(cloudErrRemoteInferenceUnavailable)+`"}` {
|
|
t.Fatalf("unexpected response: %s", got)
|
|
}
|
|
})
|
|
|
|
t.Run("show remote blocked", func(t *testing.T) {
|
|
w := createRequest(t, s.ShowHandler, api.ShowRequest{
|
|
Model: "test-cloud",
|
|
})
|
|
if w.Code != http.StatusForbidden {
|
|
t.Fatalf("expected status 403, got %d", w.Code)
|
|
}
|
|
if got := w.Body.String(); got != `{"error":"`+internalcloud.DisabledError(cloudErrRemoteModelDetailsUnavailable)+`"}` {
|
|
t.Fatalf("unexpected response: %s", got)
|
|
}
|
|
})
|
|
}
|
|
|
|
func TestDeleteHandlerNormalizesExplicitSourceSuffixes(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
s := Server{}
|
|
|
|
tests := []string{
|
|
"gpt-oss:20b:local",
|
|
"gpt-oss:20b:cloud",
|
|
"qwen3:cloud",
|
|
}
|
|
|
|
for _, modelName := range tests {
|
|
t.Run(modelName, func(t *testing.T) {
|
|
w := createRequest(t, s.DeleteHandler, api.DeleteRequest{
|
|
Model: modelName,
|
|
})
|
|
if w.Code != http.StatusNotFound {
|
|
t.Fatalf("expected status 404, got %d (%s)", w.Code, w.Body.String())
|
|
}
|
|
|
|
var resp map[string]string
|
|
if err := json.NewDecoder(w.Body).Decode(&resp); err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
want := "model '" + modelName + "' not found"
|
|
if resp["error"] != want {
|
|
t.Fatalf("unexpected error: got %q, want %q", resp["error"], want)
|
|
}
|
|
})
|
|
}
|
|
}
|
|
|
|
func TestExplicitCloudPassthroughAPIAndV1(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
type upstreamCapture struct {
|
|
path string
|
|
body string
|
|
header http.Header
|
|
}
|
|
|
|
newUpstream := func(t *testing.T, responseBody string) (*httptest.Server, *upstreamCapture) {
|
|
t.Helper()
|
|
capture := &upstreamCapture{}
|
|
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
|
payload, _ := io.ReadAll(r.Body)
|
|
capture.path = r.URL.Path
|
|
capture.body = string(payload)
|
|
capture.header = r.Header.Clone()
|
|
w.Header().Set("Content-Type", "application/json")
|
|
w.WriteHeader(http.StatusOK)
|
|
_, _ = w.Write([]byte(responseBody))
|
|
}))
|
|
|
|
return srv, capture
|
|
}
|
|
|
|
t.Run("api generate", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"ok":"api"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","prompt":"hello","stream":false}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/generate", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
req.Header.Set("X-Test-Header", "api-header")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/generate" {
|
|
t.Fatalf("expected upstream path /api/generate, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if got := capture.header.Get("X-Test-Header"); got != "api-header" {
|
|
t.Fatalf("expected forwarded X-Test-Header=api-header, got %q", got)
|
|
}
|
|
})
|
|
|
|
t.Run("api chat", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"message":{"role":"assistant","content":"ok"},"done":true}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","messages":[{"role":"user","content":"hello"}],"stream":false}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/chat", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/chat" {
|
|
t.Fatalf("expected upstream path /api/chat, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("api embed", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"model":"kimi-k2.5:cloud","embeddings":[[0.1,0.2]]}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","input":"hello"}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/embed", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/embed" {
|
|
t.Fatalf("expected upstream path /api/embed, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("api embeddings", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"embedding":[0.1,0.2]}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","prompt":"hello"}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/embeddings", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/embeddings" {
|
|
t.Fatalf("expected upstream path /api/embeddings, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("api show", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"details":{"format":"gguf"}}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud"}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/show", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/show" {
|
|
t.Fatalf("expected upstream path /api/show, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 chat completions bypasses conversion", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"chatcmpl_test","object":"chat.completion"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"gpt-oss:120b:cloud","messages":[{"role":"user","content":"hi"}],"max_tokens":7}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/chat/completions", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
req.Header.Set("X-Test-Header", "v1-header")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/chat/completions" {
|
|
t.Fatalf("expected upstream path /v1/chat/completions, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"max_tokens":7`) {
|
|
t.Fatalf("expected original OpenAI request body, got %q", capture.body)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"gpt-oss:120b"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if strings.Contains(capture.body, `"options"`) {
|
|
t.Fatalf("expected no converted Ollama options in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if got := capture.header.Get("X-Test-Header"); got != "v1-header" {
|
|
t.Fatalf("expected forwarded X-Test-Header=v1-header, got %q", got)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 chat completions bypasses conversion with legacy cloud suffix", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"chatcmpl_test","object":"chat.completion"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"gpt-oss:120b-cloud","messages":[{"role":"user","content":"hi"}],"max_tokens":7}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/chat/completions", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
req.Header.Set("X-Test-Header", "v1-legacy-header")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/chat/completions" {
|
|
t.Fatalf("expected upstream path /v1/chat/completions, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"max_tokens":7`) {
|
|
t.Fatalf("expected original OpenAI request body, got %q", capture.body)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"gpt-oss:120b"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if strings.Contains(capture.body, `"options"`) {
|
|
t.Fatalf("expected no converted Ollama options in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if got := capture.header.Get("X-Test-Header"); got != "v1-legacy-header" {
|
|
t.Fatalf("expected forwarded X-Test-Header=v1-legacy-header, got %q", got)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 messages bypasses conversion", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"msg_1","type":"message"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","max_tokens":10,"messages":[{"role":"user","content":"hi"}]}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/messages", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/messages" {
|
|
t.Fatalf("expected upstream path /v1/messages, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"max_tokens":10`) {
|
|
t.Fatalf("expected original Anthropic request body, got %q", capture.body)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if strings.Contains(capture.body, `"options"`) {
|
|
t.Fatalf("expected no converted Ollama options in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 messages bypasses conversion with legacy cloud suffix", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"msg_1","type":"message"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:latest-cloud","max_tokens":10,"messages":[{"role":"user","content":"hi"}]}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/messages", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/messages" {
|
|
t.Fatalf("expected upstream path /v1/messages, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"max_tokens":10`) {
|
|
t.Fatalf("expected original Anthropic request body, got %q", capture.body)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"kimi-k2.5:latest"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if strings.Contains(capture.body, `"options"`) {
|
|
t.Fatalf("expected no converted Ollama options in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 messages web_search fallback uses legacy cloud /api/chat path", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"model":"gpt-oss:120b","created_at":"2024-01-01T00:00:00Z","message":{"role":"assistant","content":"hello"},"done":true}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{
|
|
"model":"gpt-oss:120b-cloud",
|
|
"max_tokens":10,
|
|
"messages":[{"role":"user","content":"search the web"}],
|
|
"tools":[{"type":"web_search_20250305","name":"web_search"}],
|
|
"stream":false
|
|
}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/messages?beta=true", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/api/chat" {
|
|
t.Fatalf("expected upstream path /api/chat for web_search fallback, got %q", capture.path)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"model":"gpt-oss:120b"`) {
|
|
t.Fatalf("expected normalized model in upstream body, got %q", capture.body)
|
|
}
|
|
|
|
if !strings.Contains(capture.body, `"num_predict":10`) {
|
|
t.Fatalf("expected converted ollama options in upstream body, got %q", capture.body)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 model retrieve bypasses conversion", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"kimi-k2.5:cloud","object":"model","created":1,"owned_by":"ollama"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodGet, local.URL+"/v1/models/kimi-k2.5:cloud", nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("X-Test-Header", "v1-model-header")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/models/kimi-k2.5" {
|
|
t.Fatalf("expected upstream path /v1/models/kimi-k2.5, got %q", capture.path)
|
|
}
|
|
|
|
if capture.body != "" {
|
|
t.Fatalf("expected empty request body, got %q", capture.body)
|
|
}
|
|
|
|
if got := capture.header.Get("X-Test-Header"); got != "v1-model-header" {
|
|
t.Fatalf("expected forwarded X-Test-Header=v1-model-header, got %q", got)
|
|
}
|
|
})
|
|
|
|
t.Run("v1 model retrieve normalizes legacy cloud suffix", func(t *testing.T) {
|
|
upstream, capture := newUpstream(t, `{"id":"kimi-k2.5:latest","object":"model","created":1,"owned_by":"ollama"}`)
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodGet, local.URL+"/v1/models/kimi-k2.5:latest-cloud", nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusOK {
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "/v1/models/kimi-k2.5:latest" {
|
|
t.Fatalf("expected upstream path /v1/models/kimi-k2.5:latest, got %q", capture.path)
|
|
}
|
|
})
|
|
}
|
|
|
|
func TestCloudDisabledBlocksExplicitCloudPassthrough(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
t.Setenv("OLLAMA_NO_CLOUD", "1")
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/chat/completions", bytes.NewBufferString(`{"model":"kimi-k2.5:cloud","messages":[{"role":"user","content":"hi"}]}`))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusForbidden {
|
|
t.Fatalf("expected status 403, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
var got map[string]string
|
|
if err := json.Unmarshal(body, &got); err != nil {
|
|
t.Fatalf("expected json error body, got: %q", string(body))
|
|
}
|
|
|
|
if got["error"] != internalcloud.DisabledError(cloudErrRemoteInferenceUnavailable) {
|
|
t.Fatalf("unexpected error message: %q", got["error"])
|
|
}
|
|
}
|
|
|
|
func TestCloudPassthroughStreamsPromptly(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
|
w.Header().Set("Content-Type", "application/x-ndjson")
|
|
flusher, ok := w.(http.Flusher)
|
|
if !ok {
|
|
t.Fatal("upstream writer is not a flusher")
|
|
}
|
|
|
|
_, _ = w.Write([]byte(`{"response":"first"}` + "\n"))
|
|
flusher.Flush()
|
|
|
|
time.Sleep(700 * time.Millisecond)
|
|
|
|
_, _ = w.Write([]byte(`{"response":"second"}` + "\n"))
|
|
flusher.Flush()
|
|
}))
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","messages":[{"role":"user","content":"hi"}],"stream":true}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/chat", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
if resp.StatusCode != http.StatusOK {
|
|
body, _ := io.ReadAll(resp.Body)
|
|
t.Fatalf("expected status 200, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
reader := bufio.NewReader(resp.Body)
|
|
|
|
start := time.Now()
|
|
firstLine, err := reader.ReadString('\n')
|
|
if err != nil {
|
|
t.Fatalf("failed reading first streamed line: %v", err)
|
|
}
|
|
if elapsed := time.Since(start); elapsed > 400*time.Millisecond {
|
|
t.Fatalf("first streamed line arrived too late (%s), likely not flushing", elapsed)
|
|
}
|
|
if !strings.Contains(firstLine, `"first"`) {
|
|
t.Fatalf("expected first line to contain first chunk, got %q", firstLine)
|
|
}
|
|
|
|
secondLine, err := reader.ReadString('\n')
|
|
if err != nil {
|
|
t.Fatalf("failed reading second streamed line: %v", err)
|
|
}
|
|
if !strings.Contains(secondLine, `"second"`) {
|
|
t.Fatalf("expected second line to contain second chunk, got %q", secondLine)
|
|
}
|
|
}
|
|
|
|
func TestCloudPassthroughSkipsAnthropicWebSearch(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
type upstreamCapture struct {
|
|
path string
|
|
}
|
|
capture := &upstreamCapture{}
|
|
upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
|
capture.path = r.URL.Path
|
|
w.Header().Set("Content-Type", "application/json")
|
|
w.WriteHeader(http.StatusOK)
|
|
_, _ = w.Write([]byte(`{"id":"msg_1","type":"message"}`))
|
|
}))
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
router := gin.New()
|
|
router.POST(
|
|
"/v1/messages",
|
|
cloudPassthroughMiddleware(cloudErrRemoteInferenceUnavailable),
|
|
middleware.AnthropicMessagesMiddleware(),
|
|
func(c *gin.Context) { c.Status(http.StatusTeapot) },
|
|
)
|
|
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{
|
|
"model":"kimi-k2.5:cloud",
|
|
"max_tokens":10,
|
|
"messages":[{"role":"user","content":"hi"}],
|
|
"tools":[{"type":"web_search_20250305","name":"web_search"}]
|
|
}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/messages", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
if resp.StatusCode != http.StatusTeapot {
|
|
body, _ := io.ReadAll(resp.Body)
|
|
t.Fatalf("expected local middleware path status %d, got %d (%s)", http.StatusTeapot, resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "" {
|
|
t.Fatalf("expected no passthrough for web_search requests, got upstream path %q", capture.path)
|
|
}
|
|
}
|
|
|
|
func TestCloudPassthroughSkipsAnthropicWebSearchLegacySuffix(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
type upstreamCapture struct {
|
|
path string
|
|
}
|
|
capture := &upstreamCapture{}
|
|
upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
|
capture.path = r.URL.Path
|
|
w.Header().Set("Content-Type", "application/json")
|
|
w.WriteHeader(http.StatusOK)
|
|
_, _ = w.Write([]byte(`{"id":"msg_1","type":"message"}`))
|
|
}))
|
|
defer upstream.Close()
|
|
|
|
original := cloudProxyBaseURL
|
|
cloudProxyBaseURL = upstream.URL
|
|
t.Cleanup(func() { cloudProxyBaseURL = original })
|
|
|
|
router := gin.New()
|
|
router.POST(
|
|
"/v1/messages",
|
|
cloudPassthroughMiddleware(cloudErrRemoteInferenceUnavailable),
|
|
middleware.AnthropicMessagesMiddleware(),
|
|
func(c *gin.Context) { c.Status(http.StatusTeapot) },
|
|
)
|
|
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{
|
|
"model":"kimi-k2.5:latest-cloud",
|
|
"max_tokens":10,
|
|
"messages":[{"role":"user","content":"hi"}],
|
|
"tools":[{"type":"web_search_20250305","name":"web_search"}]
|
|
}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/v1/messages", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
if resp.StatusCode != http.StatusTeapot {
|
|
body, _ := io.ReadAll(resp.Body)
|
|
t.Fatalf("expected local middleware path status %d, got %d (%s)", http.StatusTeapot, resp.StatusCode, string(body))
|
|
}
|
|
|
|
if capture.path != "" {
|
|
t.Fatalf("expected no passthrough for web_search requests, got upstream path %q", capture.path)
|
|
}
|
|
}
|
|
|
|
func TestCloudPassthroughSigningFailureReturnsUnauthorized(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
origSignRequest := cloudProxySignRequest
|
|
origSigninURL := cloudProxySigninURL
|
|
cloudProxySignRequest = func(context.Context, *http.Request) error {
|
|
return errors.New("ssh: no key found")
|
|
}
|
|
cloudProxySigninURL = func() (string, error) {
|
|
return "https://ollama.com/signin/example", nil
|
|
}
|
|
t.Cleanup(func() {
|
|
cloudProxySignRequest = origSignRequest
|
|
cloudProxySigninURL = origSigninURL
|
|
})
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","prompt":"hello","stream":false}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/generate", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusUnauthorized {
|
|
t.Fatalf("expected status 401, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
var got map[string]any
|
|
if err := json.Unmarshal(body, &got); err != nil {
|
|
t.Fatalf("expected json error body, got: %q", string(body))
|
|
}
|
|
|
|
if got["error"] != "unauthorized" {
|
|
t.Fatalf("unexpected error message: %v", got["error"])
|
|
}
|
|
|
|
if got["signin_url"] != "https://ollama.com/signin/example" {
|
|
t.Fatalf("unexpected signin_url: %v", got["signin_url"])
|
|
}
|
|
}
|
|
|
|
func TestCloudPassthroughSigningFailureWithoutSigninURL(t *testing.T) {
|
|
gin.SetMode(gin.TestMode)
|
|
setTestHome(t, t.TempDir())
|
|
|
|
origSignRequest := cloudProxySignRequest
|
|
origSigninURL := cloudProxySigninURL
|
|
cloudProxySignRequest = func(context.Context, *http.Request) error {
|
|
return errors.New("ssh: no key found")
|
|
}
|
|
cloudProxySigninURL = func() (string, error) {
|
|
return "", errors.New("key missing")
|
|
}
|
|
t.Cleanup(func() {
|
|
cloudProxySignRequest = origSignRequest
|
|
cloudProxySigninURL = origSigninURL
|
|
})
|
|
|
|
s := &Server{}
|
|
router, err := s.GenerateRoutes(nil)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
|
|
local := httptest.NewServer(router)
|
|
defer local.Close()
|
|
|
|
reqBody := `{"model":"kimi-k2.5:cloud","prompt":"hello","stream":false}`
|
|
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost, local.URL+"/api/generate", bytes.NewBufferString(reqBody))
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
req.Header.Set("Content-Type", "application/json")
|
|
|
|
resp, err := local.Client().Do(req)
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
defer resp.Body.Close()
|
|
|
|
body, _ := io.ReadAll(resp.Body)
|
|
if resp.StatusCode != http.StatusUnauthorized {
|
|
t.Fatalf("expected status 401, got %d (%s)", resp.StatusCode, string(body))
|
|
}
|
|
|
|
var got map[string]any
|
|
if err := json.Unmarshal(body, &got); err != nil {
|
|
t.Fatalf("expected json error body, got: %q", string(body))
|
|
}
|
|
|
|
if got["error"] != "unauthorized" {
|
|
t.Fatalf("unexpected error message: %v", got["error"])
|
|
}
|
|
|
|
if _, ok := got["signin_url"]; ok {
|
|
t.Fatalf("did not expect signin_url when helper fails, got %v", got["signin_url"])
|
|
}
|
|
}
|