feat(llm): anthropic vision provider (audio unsupported)

This commit is contained in:
2026-06-07 13:07:57 +01:00
parent 967b28769f
commit f6def2f71d
2 changed files with 140 additions and 0 deletions
+91
View File
@@ -0,0 +1,91 @@
package llm
import (
"bytes"
"context"
"encoding/base64"
"encoding/json"
"fmt"
"io"
"net/http"
)
type anthropicClient struct {
http *http.Client
base string
key string
model string
maxTokens int
}
func (c *anthropicClient) Vision(ctx context.Context, prompt string, images []Image, opts Options) (string, error) {
content := []map[string]any{}
for _, img := range images {
content = append(content, map[string]any{
"type": "image",
"source": map[string]any{
"type": "base64",
"media_type": img.MediaType,
"data": base64.StdEncoding.EncodeToString(img.Data),
},
})
}
content = append(content, map[string]any{"type": "text", "text": prompt})
model := c.model
if opts.Model != "" {
model = opts.Model
}
maxTok := c.maxTokens
if opts.MaxTokens > 0 {
maxTok = opts.MaxTokens
}
// NOTE: temperature/top_p are intentionally omitted — Opus 4.x rejects them (HTTP 400).
reqBody := map[string]any{
"model": model,
"max_tokens": maxTok,
"messages": []map[string]any{{"role": "user", "content": content}},
}
b, err := json.Marshal(reqBody)
if err != nil {
return "", err
}
req, err := http.NewRequestWithContext(ctx, http.MethodPost, c.base+"/v1/messages", bytes.NewReader(b))
if err != nil {
return "", err
}
req.Header.Set("Content-Type", "application/json")
req.Header.Set("x-api-key", c.key)
req.Header.Set("anthropic-version", "2023-06-01")
resp, err := c.http.Do(req)
if err != nil {
return "", err
}
defer func() { _ = resp.Body.Close() }()
raw, err := io.ReadAll(resp.Body)
if err != nil {
return "", err
}
if resp.StatusCode >= 300 {
return "", fmt.Errorf("anthropic HTTP %d: %s", resp.StatusCode, string(raw))
}
var out struct {
Content []struct {
Type string `json:"type"`
Text string `json:"text"`
} `json:"content"`
}
if err := json.Unmarshal(raw, &out); err != nil {
return "", err
}
for _, blk := range out.Content {
if blk.Type == "text" {
return blk.Text, nil
}
}
return "", fmt.Errorf("anthropic: no text block in response")
}
func (c *anthropicClient) Transcribe(_ context.Context, _ []byte, _ string, _ Options) (string, error) {
return "", ErrUnsupported{Provider: "anthropic"}
}
+49
View File
@@ -0,0 +1,49 @@
package llm
import (
"context"
"encoding/json"
"errors"
"io"
"net/http"
"net/http/httptest"
"strings"
"testing"
"time"
)
func TestAnthropicVision(t *testing.T) {
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if r.URL.Path != "/v1/messages" {
t.Errorf("path %s", r.URL.Path)
}
if r.Header.Get("x-api-key") != "key" || r.Header.Get("anthropic-version") != "2023-06-01" {
t.Errorf("headers %v", r.Header)
}
raw, _ := io.ReadAll(r.Body)
if strings.Contains(string(raw), "temperature") {
t.Errorf("temperature must NOT be sent to anthropic: %s", raw)
}
if !strings.Contains(string(raw), `"type":"base64"`) {
t.Errorf("image block missing: %s", raw)
}
_, _ = io.WriteString(w, `{"content":[{"type":"text","text":"crosswalk"}]}`)
}))
defer srv.Close()
c := &anthropicClient{http: &http.Client{Timeout: 5 * time.Second}, base: srv.URL, key: "key", model: "claude-opus-4-8", maxTokens: 100}
got, err := c.Vision(context.Background(), "which tiles", []Image{{MediaType: "image/png", Data: []byte{2}}}, Options{Temperature: 0.7})
if err != nil || got != "crosswalk" {
t.Fatalf("got %q err %v", got, err)
}
_ = json.Marshal // keep import
}
func TestAnthropicTranscribeUnsupported(t *testing.T) {
c := &anthropicClient{model: "claude-opus-4-8"}
_, err := c.Transcribe(context.Background(), []byte{1}, "a.mp3", Options{})
var u ErrUnsupported
if !errors.As(err, &u) {
t.Fatalf("want ErrUnsupported, got %v", err)
}
}