llmshim 0.2.0

Blazing fast LLM API translation layer in pure Rust
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272

use llmshim::provider::Provider;
use llmshim::providers::anthropic::Anthropic;
use llmshim::providers::gemini::Gemini;
use llmshim::providers::openai::OpenAi;
use llmshim::providers::xai::Xai;
use serde_json::json;

// ============================================================
// Anthropic — fast mode
// ============================================================

#[test]
fn anthropic_fast_mode_adds_speed_to_body() {
    let p = Anthropic::new("test-key".into());
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    assert_eq!(result.body["speed"], "fast");
}

#[test]
fn anthropic_fast_mode_adds_beta_header() {
    let p = Anthropic::new("test-key".into());
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    let beta_header = result
        .headers
        .iter()
        .find(|(k, _)| k == "anthropic-beta")
        .expect("Expected anthropic-beta header");
    assert!(
        beta_header.1.contains("fast-mode-2026-02-01"),
        "Beta header should contain fast-mode-2026-02-01, got: {}",
        beta_header.1
    );
}

#[test]
fn anthropic_fast_mode_combines_beta_headers_with_1m_context() {
    let p = Anthropic::new("test-key".into());
    // claude-opus-4-6 supports 1M context, so both betas should be present
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    let beta_header = result
        .headers
        .iter()
        .find(|(k, _)| k == "anthropic-beta")
        .expect("Expected anthropic-beta header");
    assert!(
        beta_header.1.contains("context-1m-2025-08-07"),
        "Beta header should contain context-1m beta, got: {}",
        beta_header.1
    );
    assert!(
        beta_header.1.contains("fast-mode-2026-02-01"),
        "Beta header should contain fast-mode beta, got: {}",
        beta_header.1
    );
}

#[test]
fn anthropic_no_speed_no_fast_mode_header() {
    let p = Anthropic::new("test-key".into());
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    let beta_header = result.headers.iter().find(|(k, _)| k == "anthropic-beta");
    if let Some((_, val)) = beta_header {
        assert!(
            !val.contains("fast-mode"),
            "Should not contain fast-mode beta without speed param, got: {}",
            val
        );
    }
}

#[test]
fn anthropic_speed_not_fast_no_fast_mode_header() {
    let p = Anthropic::new("test-key".into());
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "standard",
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    // "speed" should still be in body (pass through)
    assert_eq!(result.body["speed"], "standard");
    // But fast-mode beta header should NOT be added
    let beta_header = result.headers.iter().find(|(k, _)| k == "anthropic-beta");
    if let Some((_, val)) = beta_header {
        assert!(
            !val.contains("fast-mode"),
            "Should not contain fast-mode beta for speed=standard, got: {}",
            val
        );
    }
}

#[test]
fn anthropic_fast_mode_works_with_non_1m_model() {
    let p = Anthropic::new("test-key".into());
    // Use a model that does NOT support 1M context, so only fast-mode beta should appear
    let req = json!({
        "model": "claude-3-5-sonnet-20241022",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p
        .transform_request("claude-3-5-sonnet-20241022", &req)
        .unwrap();
    assert_eq!(result.body["speed"], "fast");
    let beta_header = result
        .headers
        .iter()
        .find(|(k, _)| k == "anthropic-beta")
        .expect("Expected anthropic-beta header for fast mode");
    assert_eq!(beta_header.1, "fast-mode-2026-02-01");
}

// ============================================================
// OpenAI — fast mode / priority processing
// ============================================================

#[test]
fn openai_fast_mode_adds_priority() {
    let p = OpenAi::new("test-key".into());
    let req = json!({
        "model": "gpt-5.4",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("gpt-5.4", &req).unwrap();
    assert_eq!(result.body["priority"]["type"], "default_with_boost");
}

#[test]
fn openai_no_speed_no_priority() {
    let p = OpenAi::new("test-key".into());
    let req = json!({
        "model": "gpt-5.4",
        "messages": [{"role": "user", "content": "hi"}],
    });
    let result = p.transform_request("gpt-5.4", &req).unwrap();
    assert!(
        result.body.get("priority").is_none(),
        "Should not have priority field without speed param"
    );
}

#[test]
fn openai_speed_standard_no_priority() {
    let p = OpenAi::new("test-key".into());
    let req = json!({
        "model": "gpt-5.4",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "standard",
    });
    let result = p.transform_request("gpt-5.4", &req).unwrap();
    assert!(
        result.body.get("priority").is_none(),
        "Should not have priority field for speed=standard"
    );
}

#[test]
fn openai_fast_mode_does_not_pass_speed_to_body() {
    let p = OpenAi::new("test-key".into());
    let req = json!({
        "model": "gpt-5.4",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("gpt-5.4", &req).unwrap();
    // OpenAI does not use "speed" field — only "priority"
    assert!(
        result.body.get("speed").is_none(),
        "OpenAI body should not contain 'speed' field"
    );
}

// ============================================================
// Gemini — fast mode silently ignored
// ============================================================

#[test]
fn gemini_fast_mode_ignored() {
    let p = Gemini::new("test-key".into());
    let req = json!({
        "model": "gemini-3-flash-preview",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("gemini-3-flash-preview", &req).unwrap();
    // "speed" should not appear in the Gemini request body
    assert!(
        result.body.get("speed").is_none(),
        "Gemini body should not contain 'speed' field"
    );
    assert!(
        result.body.get("priority").is_none(),
        "Gemini body should not contain 'priority' field"
    );
}

// ============================================================
// xAI — fast mode silently ignored
// ============================================================

#[test]
fn xai_fast_mode_ignored() {
    let p = Xai::new("test-key".into());
    let req = json!({
        "model": "grok-4.3",
        "messages": [{"role": "user", "content": "hi"}],
        "speed": "fast",
    });
    let result = p.transform_request("grok-4.3", &req).unwrap();
    // "speed" should not appear in the xAI request body
    assert!(
        result.body.get("speed").is_none(),
        "xAI body should not contain 'speed' field"
    );
    assert!(
        result.body.get("priority").is_none(),
        "xAI body should not contain 'priority' field"
    );
}

// ============================================================
// Backward compatibility — existing callers work unchanged
// ============================================================

#[test]
fn anthropic_request_without_speed_unchanged() {
    let p = Anthropic::new("test-key".into());
    let req = json!({
        "model": "claude-opus-4-6",
        "messages": [{"role": "user", "content": "hi"}],
        "max_tokens": 1024,
    });
    let result = p.transform_request("claude-opus-4-6", &req).unwrap();
    assert!(result.body.get("speed").is_none());
    assert_eq!(result.body["max_tokens"], 1024);
    assert_eq!(result.body["model"], "claude-opus-4-6");
}

#[test]
fn openai_request_without_speed_unchanged() {
    let p = OpenAi::new("test-key".into());
    let req = json!({
        "model": "gpt-5.4",
        "messages": [{"role": "user", "content": "hi"}],
        "max_tokens": 1024,
    });
    let result = p.transform_request("gpt-5.4", &req).unwrap();
    assert!(result.body.get("speed").is_none());
    assert!(result.body.get("priority").is_none());
    assert_eq!(result.body["model"], "gpt-5.4");
}