tt-shared 0.1.1

Shared types, errors, and Provider trait for TokenTrimmer.
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

# Model metadata catalog — per-(provider, model) context windows + capabilities.
# RATES live in pricing.toml; this file is METADATA ONLY. Embedded at build time
# (see model_catalog.rs) and parsed once into the ModelCatalog — the single
# source of truth for ModelInfo across provider adapters and GET /v1/models.
#
# Schema — one [[model]] per (provider, model):
#   provider          = registry provider id ("openai", "anthropic", "gemini", …)
#   model             = exact model id the provider matches on
#   max_input_tokens  = context window (input) upper bound
#   max_output_tokens = completion upper bound (0 for embedding models)
#   capabilities      = snake_case Capability names: text, vision, audio, tools,
#                       json_mode, streaming, reasoning, prompt_caching
#
# V4b-1 covers the native adapters (openai, anthropic, gemini); compat providers
# (mistral, groq, together, openrouter) are added in V4b-2.

# ── OpenAI ──────────────────────────────────────────────────────────────────
[[model]]
provider = "openai"
model = "gpt-5.5"
max_input_tokens = 200000
max_output_tokens = 16000
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "openai"
model = "gpt-5.4"
max_input_tokens = 200000
max_output_tokens = 16000
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "openai"
model = "gpt-4o"
max_input_tokens = 128000
max_output_tokens = 16000
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "openai"
model = "gpt-4o-mini"
max_input_tokens = 128000
max_output_tokens = 16000
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "openai"
model = "o3"
max_input_tokens = 200000
max_output_tokens = 100000
capabilities = ["text", "tools", "json_mode", "reasoning", "streaming"]

[[model]]
provider = "openai"
model = "o4-mini"
max_input_tokens = 200000
max_output_tokens = 100000
capabilities = ["text", "tools", "json_mode", "reasoning", "streaming"]

[[model]]
provider = "openai"
model = "text-embedding-3-small"
max_input_tokens = 8191
max_output_tokens = 0
capabilities = ["text"]

[[model]]
provider = "openai"
model = "text-embedding-3-large"
max_input_tokens = 8191
max_output_tokens = 0
capabilities = ["text"]

# ── Anthropic ───────────────────────────────────────────────────────────────
[[model]]
provider = "anthropic"
model = "claude-haiku-4-5"
max_input_tokens = 200000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "anthropic"
model = "claude-sonnet-4-6"
max_input_tokens = 200000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "anthropic"
model = "claude-opus-4-7"
max_input_tokens = 200000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

# ── Gemini ──────────────────────────────────────────────────────────────────
[[model]]
provider = "gemini"
model = "gemini-3.1-flash-lite"
max_input_tokens = 1000000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "gemini"
model = "gemini-3.5-flash"
max_input_tokens = 1000000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

[[model]]
provider = "gemini"
model = "gemini-3.1-pro"
max_input_tokens = 2000000
max_output_tokens = 8192
capabilities = ["text", "vision", "tools", "json_mode", "streaming", "prompt_caching"]

# ── Mistral (compat) ──────────────────────────────────────────────────────────
[[model]]
provider = "mistral"
model = "mistral-large-latest"
max_input_tokens = 128000
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "mistral"
model = "mistral-medium-latest"
max_input_tokens = 128000
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "mistral"
model = "mistral-small-latest"
max_input_tokens = 128000
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "mistral"
model = "codestral-latest"
max_input_tokens = 256000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "mistral"
model = "pixtral-large-latest"
max_input_tokens = 128000
max_output_tokens = 4096
capabilities = ["text", "vision", "tools", "json_mode", "streaming"]

# ── Groq (compat) ─────────────────────────────────────────────────────────────
[[model]]
provider = "groq"
model = "llama-3.3-70b-versatile"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "groq"
model = "llama-3.1-8b-instant"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "groq"
model = "deepseek-r1-distill-llama-70b"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming", "reasoning"]

[[model]]
provider = "groq"
model = "mixtral-8x7b-32768"
max_input_tokens = 32768
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]

# ── Together (compat) ─────────────────────────────────────────────────────────
[[model]]
provider = "together"
model = "meta-llama/Meta-Llama-3.3-70B-Instruct-Turbo"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "together"
model = "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "together"
model = "Qwen/Qwen2.5-72B-Instruct-Turbo"
max_input_tokens = 32768
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "together"
model = "deepseek-ai/DeepSeek-V3"
max_input_tokens = 64000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

# ── OpenRouter (compat; provider-namespaced ids) ──────────────────────────────
[[model]]
provider = "openrouter"
model = "anthropic/claude-sonnet-4-6"
max_input_tokens = 200000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "openrouter"
model = "openai/gpt-5.5"
max_input_tokens = 200000
max_output_tokens = 16000
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "openrouter"
model = "google/gemini-3.1-pro"
max_input_tokens = 1000000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "openrouter"
model = "meta-llama/llama-3.3-70b-instruct"
max_input_tokens = 128000
max_output_tokens = 8192
capabilities = ["text", "tools", "json_mode", "streaming"]

[[model]]
provider = "openrouter"
model = "mistralai/mistral-large"
max_input_tokens = 128000
max_output_tokens = 4096
capabilities = ["text", "tools", "json_mode", "streaming"]