or run

npx @tessl/cli init
Log in

Version

Tile

Overview

Evals

Files

Files

docs

audio.mdbatches.mdchat-completions.mdembeddings.mdfiles.mdindex.mdmodels.md

models.mddocs/

0

# Model Management

1

2

Access model information, list available models, and manage model lifecycle operations. The models API provides comprehensive information about available models, their capabilities, and management functions.

3

4

## Capabilities

5

6

### List Models

7

8

Retrieve a list of all available models with their metadata and capabilities.

9

10

```python { .api }

11

def list(

12

extra_headers: Headers | None = None,

13

extra_query: Query | None = None,

14

extra_body: Body | None = None,

15

timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN

16

) -> ModelListResponse:

17

"""

18

List all available models.

19

20

Returns:

21

ModelListResponse containing list of available models with metadata

22

"""

23

```

24

25

### Retrieve Model

26

27

Get detailed information about a specific model by its identifier.

28

29

```python { .api }

30

def retrieve(

31

model: str,

32

extra_headers: Headers | None = None,

33

extra_query: Query | None = None,

34

extra_body: Body | None = None,

35

timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN

36

) -> Model:

37

"""

38

Retrieve information about a specific model.

39

40

Parameters:

41

- model: Model identifier to retrieve information for

42

43

Returns:

44

Model object with detailed information about the specified model

45

"""

46

```

47

48

### Delete Model

49

50

Remove a model from your account (for fine-tuned or custom models).

51

52

```python { .api }

53

def delete(

54

model: str,

55

extra_headers: Headers | None = None,

56

extra_query: Query | None = None,

57

extra_body: Body | None = None,

58

timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN

59

) -> ModelDeleted:

60

"""

61

Delete a model.

62

63

Parameters:

64

- model: Model identifier to delete

65

66

Returns:

67

ModelDeleted confirmation object

68

"""

69

```

70

71

### Async Model Operations

72

73

All model operations have asynchronous counterparts with identical parameters.

74

75

```python { .api }

76

async def list(**kwargs) -> ModelListResponse: ...

77

async def retrieve(model: str, **kwargs) -> Model: ...

78

async def delete(model: str, **kwargs) -> ModelDeleted: ...

79

```

80

81

## Usage Examples

82

83

### List Available Models

84

85

```python

86

from groq import Groq

87

88

client = Groq()

89

90

# Get all available models

91

models = client.models.list()

92

93

print(f"Available models: {len(models.data)}")

94

for model in models.data:

95

print(f"- {model.id}: {model.object}")

96

if hasattr(model, 'owned_by'):

97

print(f" Owned by: {model.owned_by}")

98

if hasattr(model, 'created'):

99

print(f" Created: {model.created}")

100

```

101

102

### Get Model Information

103

104

```python

105

from groq import Groq

106

107

client = Groq()

108

109

# Get information about a specific model

110

model = client.models.retrieve("llama3-8b-8192")

111

112

print(f"Model ID: {model.id}")

113

print(f"Object type: {model.object}")

114

print(f"Created: {model.created}")

115

print(f"Owned by: {model.owned_by}")

116

```

117

118

### Check Model Capabilities

119

120

```python

121

from groq import Groq

122

123

client = Groq()

124

125

# List models and check their capabilities

126

models = client.models.list()

127

128

chat_models = []

129

embedding_models = []

130

131

for model in models.data:

132

model_info = client.models.retrieve(model.id)

133

134

# Categorize models based on their ID patterns

135

if "embed" in model.id.lower():

136

embedding_models.append(model.id)

137

else:

138

chat_models.append(model.id)

139

140

print("Chat/Completion Models:")

141

for model_id in chat_models:

142

print(f" - {model_id}")

143

144

print("\nEmbedding Models:")

145

for model_id in embedding_models:

146

print(f" - {model_id}")

147

```

148

149

### Async Usage

150

151

```python

152

import asyncio

153

from groq import AsyncGroq

154

155

async def main():

156

client = AsyncGroq()

157

158

# Async model listing

159

models = await client.models.list()

160

print(f"Found {len(models.data)} models")

161

162

# Async model retrieval

163

if models.data:

164

first_model = await client.models.retrieve(models.data[0].id)

165

print(f"First model: {first_model.id}")

166

167

asyncio.run(main())

168

```

169

170

### Delete Custom Model

171

172

```python

173

from groq import Groq

174

175

client = Groq()

176

177

# Delete a custom or fine-tuned model

178

try:

179

result = client.models.delete("ft:llama3-8b-8192:my-org:custom-model:abc123")

180

print(f"Model deleted: {result.deleted}")

181

print(f"Model ID: {result.id}")

182

except Exception as e:

183

print(f"Failed to delete model: {e}")

184

```

185

186

## Types

187

188

### Model Information Types

189

190

```python { .api }

191

class Model:

192

id: str

193

object: Literal["model"]

194

created: int

195

owned_by: str

196

197

class ModelListResponse:

198

object: Literal["list"]

199

data: List[Model]

200

201

class ModelDeleted:

202

id: str

203

object: Literal["model"]

204

deleted: bool

205

```

206

207

### Common Model Identifiers

208

209

Popular models available through the Groq API:

210

211

#### Chat/Completion Models

212

- `llama3-8b-8192` - Llama 3 8B model with 8192 context length

213

- `llama3-70b-8192` - Llama 3 70B model with 8192 context length

214

- `mixtral-8x7b-32768` - Mixtral 8x7B model with 32768 context length

215

- `gemma-7b-it` - Gemma 7B instruction-tuned model

216

217

#### Audio Models

218

- `whisper-large-v3` - Whisper large v3 for transcription and translation

219

- `tts-1` - Text-to-speech model

220

- `tts-1-hd` - High-definition text-to-speech model

221

222

#### Embedding Models

223

- `nomic-embed-text-v1_5` - Nomic text embedding model v1.5