lm-deluge 0.0.35__py3-none-any.whl → 0.0.37__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of lm-deluge might be problematic. Click here for more details.

Files changed (28) hide show
  1. lm_deluge/api_requests/anthropic.py +3 -3
  2. lm_deluge/api_requests/gemini.py +3 -1
  3. lm_deluge/api_requests/openai.py +3 -1
  4. lm_deluge/api_requests/response.py +4 -3
  5. lm_deluge/cli.py +300 -0
  6. lm_deluge/client.py +7 -4
  7. lm_deluge/models/__init__.py +36 -1282
  8. lm_deluge/models/anthropic.py +124 -0
  9. lm_deluge/models/bedrock.py +99 -0
  10. lm_deluge/models/cerebras.py +57 -0
  11. lm_deluge/models/cohere.py +98 -0
  12. lm_deluge/models/deepseek.py +27 -0
  13. lm_deluge/models/fireworks.py +16 -0
  14. lm_deluge/models/google.py +153 -0
  15. lm_deluge/models/grok.py +38 -0
  16. lm_deluge/models/groq.py +74 -0
  17. lm_deluge/models/meta.py +65 -0
  18. lm_deluge/models/mistral.py +110 -0
  19. lm_deluge/models/openai.py +318 -0
  20. lm_deluge/models/openrouter.py +1 -0
  21. lm_deluge/models/together.py +112 -0
  22. lm_deluge/prompt.py +2 -2
  23. lm_deluge/util/harmony.py +6 -4
  24. {lm_deluge-0.0.35.dist-info → lm_deluge-0.0.37.dist-info}/METADATA +1 -1
  25. {lm_deluge-0.0.35.dist-info → lm_deluge-0.0.37.dist-info}/RECORD +28 -13
  26. {lm_deluge-0.0.35.dist-info → lm_deluge-0.0.37.dist-info}/WHEEL +0 -0
  27. {lm_deluge-0.0.35.dist-info → lm_deluge-0.0.37.dist-info}/licenses/LICENSE +0 -0
  28. {lm_deluge-0.0.35.dist-info → lm_deluge-0.0.37.dist-info}/top_level.txt +0 -0
lm_deluge/models/anthropic.py ADDED
@@ -0,0 +1,124 @@
1
+ ANTHROPIC_MODELS = {
2
+ # █████████ █████ █████ ███
3
+ # ███░░░░░███ ░░███ ░░███ ░░░
4
+ # ░███ ░███ ████████ ███████ ░███████ ████████ ██████ ████████ ████ ██████
5
+ # ░███████████ ░░███░░███ ░░░███░ ░███░░███ ░░███░░███ ███░░███░░███░░███░░███ ███░░███
6
+ # ░███░░░░░███ ░███ ░███ ░███ ░███ ░███ ░███ ░░░ ░███ ░███ ░███ ░███ ░███ ░███ ░░░
7
+ # ░███ ░███ ░███ ░███ ░███ ███ ░███ ░███ ░███ ░███ ░███ ░███ ░███ ░███ ░███ ███
8
+ # █████ █████ ████ █████ ░░█████ ████ █████ █████ ░░██████ ░███████ █████░░██████
9
+ # ░░░░░ ░░░░░ ░░░░ ░░░░░ ░░░░░ ░░░░ ░░░░░ ░░░░░ ░░░░░░ ░███░░░ ░░░░░ ░░░░░░
10
+ # ░███
11
+ # █████
12
+ #
13
+ "claude-4.1-opus": {
14
+ "id": "claude-4.1-opus",
15
+ "name": "claude-opus-4-1-20250805",
16
+ "api_base": "https://api.anthropic.com/v1",
17
+ "api_key_env_var": "ANTHROPIC_API_KEY",
18
+ "supports_json": False,
19
+ "api_spec": "anthropic",
20
+ "input_cost": 15.0,
21
+ "output_cost": 75.0,
22
+ "requests_per_minute": 4_000,
23
+ "tokens_per_minute": 400_000,
24
+ "reasoning_model": True,
25
+ },
26
+ "claude-4-opus": {
27
+ "id": "claude-4-opus",
28
+ "name": "claude-opus-4-20250514",
29
+ "api_base": "https://api.anthropic.com/v1",
30
+ "api_key_env_var": "ANTHROPIC_API_KEY",
31
+ "supports_json": False,
32
+ "api_spec": "anthropic",
33
+ "input_cost": 15.0,
34
+ "output_cost": 75.0,
35
+ "requests_per_minute": 4_000,
36
+ "tokens_per_minute": 400_000,
37
+ "reasoning_model": True,
38
+ },
39
+ "claude-4-sonnet": {
40
+ "id": "claude-4-sonnet",
41
+ "name": "claude-sonnet-4-20250514",
42
+ "api_base": "https://api.anthropic.com/v1",
43
+ "api_key_env_var": "ANTHROPIC_API_KEY",
44
+ "supports_json": False,
45
+ "api_spec": "anthropic",
46
+ "input_cost": 3.0,
47
+ "output_cost": 15.0,
48
+ "requests_per_minute": 4_000,
49
+ "tokens_per_minute": 400_000,
50
+ },
51
+ "claude-3.7-sonnet": {
52
+ "id": "claude-3.7-sonnet",
53
+ "name": "claude-3-7-sonnet-20250219",
54
+ "api_base": "https://api.anthropic.com/v1",
55
+ "api_key_env_var": "ANTHROPIC_API_KEY",
56
+ "supports_json": False,
57
+ "api_spec": "anthropic",
58
+ "input_cost": 3.0,
59
+ "output_cost": 15.0,
60
+ "requests_per_minute": 4_000,
61
+ "tokens_per_minute": 400_000,
62
+ "reasoning_model": True,
63
+ },
64
+ "claude-3.6-sonnet": {
65
+ "id": "claude-3.6-sonnet",
66
+ "name": "claude-3-5-sonnet-20241022",
67
+ "api_base": "https://api.anthropic.com/v1",
68
+ "api_key_env_var": "ANTHROPIC_API_KEY",
69
+ "supports_json": False,
70
+ "api_spec": "anthropic",
71
+ "input_cost": 3.0,
72
+ "output_cost": 15.0,
73
+ "requests_per_minute": 4_000,
74
+ "tokens_per_minute": 400_000,
75
+ },
76
+ "claude-3.5-sonnet": {
77
+ "id": "claude-3.5-sonnet",
78
+ "name": "claude-3-5-sonnet-20240620",
79
+ "api_base": "https://api.anthropic.com/v1",
80
+ "api_key_env_var": "ANTHROPIC_API_KEY",
81
+ "supports_json": False,
82
+ "api_spec": "anthropic",
83
+ "input_cost": 3.0,
84
+ "output_cost": 15.0,
85
+ "requests_per_minute": 4_000,
86
+ "tokens_per_minute": 400_000,
87
+ },
88
+ "claude-3-opus": {
89
+ "id": "claude-3-opus",
90
+ "name": "claude-3-opus-20240229",
91
+ "api_base": "https://api.anthropic.com/v1",
92
+ "api_key_env_var": "ANTHROPIC_API_KEY",
93
+ "supports_json": False,
94
+ "api_spec": "anthropic",
95
+ "input_cost": 15.0,
96
+ "output_cost": 75.0,
97
+ "requests_per_minute": 4_000,
98
+ "tokens_per_minute": 400_000,
99
+ },
100
+ "claude-3.5-haiku": {
101
+ "id": "claude-3.5-haiku",
102
+ "name": "claude-3-5-haiku-20241022",
103
+ "api_base": "https://api.anthropic.com/v1",
104
+ "api_key_env_var": "ANTHROPIC_API_KEY",
105
+ "supports_json": False,
106
+ "api_spec": "anthropic",
107
+ "input_cost": 1.00,
108
+ "output_cost": 5.00,
109
+ "requests_per_minute": 20_000,
110
+ "tokens_per_minute": 4_000_000, # supposed to be this but they fucked up
111
+ },
112
+ "claude-3-haiku": {
113
+ "id": "claude-3-haiku",
114
+ "name": "claude-3-haiku-20240307",
115
+ "api_base": "https://api.anthropic.com/v1",
116
+ "api_key_env_var": "ANTHROPIC_API_KEY",
117
+ "supports_json": False,
118
+ "api_spec": "anthropic",
119
+ "input_cost": 0.25,
120
+ "output_cost": 1.25,
121
+ "requests_per_minute": 10_000,
122
+ "tokens_per_minute": 4_000_000, # supposed to be this but they fucked up
123
+ },
124
+ }
lm_deluge/models/bedrock.py ADDED
@@ -0,0 +1,99 @@
1
+ BEDROCK_MODELS = {
2
+ # ███████████ █████ █████
3
+ # ░░███░░░░░███ ░░███ ░░███
4
+ # ░███ ░███ ██████ ███████ ████████ ██████ ██████ ░███ █████
5
+ # ░██████████ ███░░███ ███░░███ ░░███░░███ ███░░███ ███░░███ ░███░░███
6
+ # ░███░░░░░███░███████ ░███ ░███ ░███ ░░░ ░███ ░███░███ ░░░ ░██████░
7
+ # ░███ ░███░███░░░ ░███ ░███ ░███ ░███ ░███░███ ███ ░███░░███
8
+ # ███████████ ░░██████ ░░████████ █████ ░░██████ ░░██████ ████ █████
9
+ # ░░░░░░░░░░░ ░░░░░░ ░░░░░░░░ ░░░░░ ░░░░░░ ░░░░░░ ░░░░ ░░░░░
10
+ "claude-3-haiku-bedrock": {
11
+ "id": "claude-3-haiku-bedrock",
12
+ "name": "us.anthropic.claude-3-haiku-20240307-v1:0",
13
+ "regions": ["us-east-1", "us-west-2", "ap-southeast-2", "eu-west-3"],
14
+ "api_base": "",
15
+ "api_key_env_var": "",
16
+ "api_spec": "bedrock",
17
+ "input_cost": 0.25,
18
+ "output_cost": 1.25,
19
+ "requests_per_minute": 4_000,
20
+ "tokens_per_minute": 8_000_000,
21
+ },
22
+ "claude-3.5-haiku-bedrock": {
23
+ "id": "claude-3.5-haiku-bedrock",
24
+ "name": "us.anthropic.claude-3-5-haiku-20241022-v1:0",
25
+ "regions": ["us-east-1", "us-west-2", "ap-southeast-2", "eu-west-3"],
26
+ "api_base": "",
27
+ "api_key_env_var": "",
28
+ "api_spec": "bedrock",
29
+ "input_cost": 0.25,
30
+ "output_cost": 1.25,
31
+ "requests_per_minute": 4_000,
32
+ "tokens_per_minute": 8_000_000,
33
+ },
34
+ "claude-3.5-sonnet-bedrock": {
35
+ "id": "claude-3.5-sonnet-bedrock",
36
+ "name": "us.anthropic.claude-3-5-sonnet-20240620-v1:0",
37
+ "regions": ["us-east-1", "us-west-2"],
38
+ "api_base": "",
39
+ "api_key_env_var": "",
40
+ "api_spec": "bedrock",
41
+ "input_cost": 3.0,
42
+ "output_cost": 15.0,
43
+ "requests_per_minute": 4_000,
44
+ "tokens_per_minute": 400_000,
45
+ "reasoning_model": False,
46
+ },
47
+ "claude-3.6-sonnet-bedrock": {
48
+ "id": "claude-3.6-sonnet-bedrock",
49
+ "name": "us.anthropic.claude-3-5-sonnet-20241022-v2:0",
50
+ "regions": ["us-east-1", "us-west-2", "us-east-2"],
51
+ "api_base": "",
52
+ "api_key_env_var": "",
53
+ "api_spec": "bedrock",
54
+ "input_cost": 3.0,
55
+ "output_cost": 15.0,
56
+ "requests_per_minute": 4_000,
57
+ "tokens_per_minute": 400_000,
58
+ "reasoning_model": False,
59
+ },
60
+ "claude-3.7-sonnet-bedrock": {
61
+ "id": "claude-3.7-sonnet-bedrock",
62
+ "name": "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
63
+ "regions": ["us-east-1", "us-west-2", "eu-west-1"],
64
+ "api_base": "",
65
+ "api_key_env_var": "",
66
+ "api_spec": "bedrock",
67
+ "input_cost": 3.0,
68
+ "output_cost": 15.0,
69
+ "requests_per_minute": 4_000,
70
+ "tokens_per_minute": 400_000,
71
+ "reasoning_model": True,
72
+ },
73
+ "claude-4-sonnet-bedrock": {
74
+ "id": "claude-4-sonnet-bedrock",
75
+ "name": "us.anthropic.claude-sonnet-4-20250514-v1:0",
76
+ "regions": ["us-east-1", "us-west-2", "us-east-2"],
77
+ "api_base": "",
78
+ "api_key_env_var": "",
79
+ "api_spec": "bedrock",
80
+ "input_cost": 3.0,
81
+ "output_cost": 15.0,
82
+ "requests_per_minute": 4_000,
83
+ "tokens_per_minute": 400_000,
84
+ "reasoning_model": True,
85
+ },
86
+ "claude-4-opus-bedrock": {
87
+ "id": "claude-4-opus-bedrock",
88
+ "name": "us.anthropic.claude-opus-4-20250514-v1:0",
89
+ "regions": ["us-east-1", "us-west-2", "us-east-2"],
90
+ "api_base": "",
91
+ "api_key_env_var": "",
92
+ "api_spec": "bedrock",
93
+ "input_cost": 3.0,
94
+ "output_cost": 15.0,
95
+ "requests_per_minute": 4_000,
96
+ "tokens_per_minute": 400_000,
97
+ "reasoning_model": True,
98
+ },
99
+ }
lm_deluge/models/cerebras.py ADDED
@@ -0,0 +1,57 @@
1
+ CEREBRAS_MODELS = {
2
+ "gpt-oss-120b-cerebras": {
3
+ "id": "gpt-oss-120b-cerebras",
4
+ "name": "gpt-oss-120b",
5
+ "api_base": "https://api.cerebras.ai/v1",
6
+ "api_key_env_var": "CEREBRAS_API_KEY",
7
+ },
8
+ "llama-4-scout-cerebras": {
9
+ "id": "llama-4-scout-cerebras",
10
+ "name": "llama-4-scout-17b-16e-instruct",
11
+ "api_base": "https://api.cerebras.ai/v1",
12
+ "api_key_env_var": "CEREBRAS_API_KEY",
13
+ },
14
+ "llama-3.1-8b-cerebras": {
15
+ "id": "llama-3.1-8b-cerebras",
16
+ "name": "llama-3.1-8b",
17
+ "api_base": "https://api.cerebras.ai/v1",
18
+ "api_key_env_var": "CEREBRAS_API_KEY",
19
+ },
20
+ "llama-3.3-70b-cerebras": {
21
+ "id": "llama-3.3-70b-cerebras",
22
+ "name": "llama-3.3-70b",
23
+ "api_base": "https://api.cerebras.ai/v1",
24
+ "api_key_env_var": "CEREBRAS_API_KEY",
25
+ },
26
+ "qwen-3-32b-cerebras": {
27
+ "id": "qwen-3-32b-cerebras",
28
+ "name": "qwen-3-32b",
29
+ "api_base": "https://api.cerebras.ai/v1",
30
+ "api_key_env_var": "CEREBRAS_API_KEY",
31
+ },
32
+ # preview models
33
+ "llama-4-maverick-cerebras": {
34
+ "id": "llama-4-maverick-cerebras",
35
+ "name": "llama-4-maverick-17b-128e-instruct",
36
+ "api_base": "https://api.cerebras.ai/v1",
37
+ "api_key_env_var": "CEREBRAS_API_KEY",
38
+ },
39
+ "qwen-3-235b-instruct-cerebras": {
40
+ "id": "qwen-3-235b-instruct-cerebras",
41
+ "name": "qwen-3-235b-a22b-instruct-2507",
42
+ "api_base": "https://api.cerebras.ai/v1",
43
+ "api_key_env_var": "CEREBRAS_API_KEY",
44
+ },
45
+ "qwen-3-235b-thinking-cerebras": {
46
+ "id": "qwen-3-235b-thinking-cerebras",
47
+ "name": "qwen-3-235b-a22b-thinking-2507",
48
+ "api_base": "https://api.cerebras.ai/v1",
49
+ "api_key_env_var": "CEREBRAS_API_KEY",
50
+ },
51
+ "qwen-3-coder-cerebras": {
52
+ "id": "qwen-3-coder-cerebras",
53
+ "name": "qwen-3-coder-480b",
54
+ "api_base": "https://api.cerebras.ai/v1",
55
+ "api_key_env_var": "CEREBRAS_API_KEY",
56
+ },
57
+ }
lm_deluge/models/cohere.py ADDED
@@ -0,0 +1,98 @@
1
+ COHERE_MODELS = {
2
+ # █████████ █████
3
+ # ███░░░░░███ ░░███
4
+ # ███ ░░░ ██████ ░███████ ██████ ████████ ██████
5
+ # ░███ ███░░███ ░███░░███ ███░░███░░███░░███ ███░░███
6
+ # ░███ ░███ ░███ ░███ ░███ ░███████ ░███ ░░░ ░███████
7
+ # ░░███ ███░███ ░███ ░███ ░███ ░███░░░ ░███ ░███░░░
8
+ # ░░█████████ ░░██████ ████ █████░░██████ █████ ░░██████
9
+ # ░░░░░░░░░ ░░░░░░ ░░░░ ░░░░░ ░░░░░░ ░░░░░ ░░░░░░
10
+ "aya-expanse-8b": {
11
+ "id": "aya-expanse-8b",
12
+ "name": "c4ai-aya-expanse-8b",
13
+ "api_base": "https://api.cohere.ai/compatibility/v1",
14
+ "api_key_env_var": "COHERE_API_KEY",
15
+ "api_spec": "openai",
16
+ "input_cost": 0.5,
17
+ "output_cost": 1.5,
18
+ "requests_per_minute": 10_000,
19
+ "tokens_per_minute": None,
20
+ },
21
+ "aya-expanse-32b": {
22
+ "id": "aya-expanse-32b",
23
+ "name": "c4ai-aya-expanse-32b",
24
+ "api_base": "https://api.cohere.ai/compatibility/v1",
25
+ "api_key_env_var": "COHERE_API_KEY",
26
+ "api_spec": "openai",
27
+ "input_cost": 0.5,
28
+ "output_cost": 1.5,
29
+ "requests_per_minute": 10_000,
30
+ "tokens_per_minute": None,
31
+ },
32
+ "aya-vision-8b": {
33
+ "id": "aya-vision-8b",
34
+ "name": "c4ai-aya-vision-8b",
35
+ "api_base": "https://api.cohere.ai/compatibility/v1",
36
+ "api_key_env_var": "COHERE_API_KEY",
37
+ "api_spec": "openai",
38
+ "input_cost": 0.5,
39
+ "output_cost": 1.5,
40
+ "requests_per_minute": 10_000,
41
+ "tokens_per_minute": None,
42
+ },
43
+ "aya-vision-32b": {
44
+ "id": "aya-vision-32b",
45
+ "name": "c4ai-aya-vision-32b",
46
+ "api_base": "https://api.cohere.ai/compatibility/v1",
47
+ "api_key_env_var": "COHERE_API_KEY",
48
+ "api_spec": "openai",
49
+ "input_cost": 0.5,
50
+ "output_cost": 1.5,
51
+ "requests_per_minute": 10_000,
52
+ "tokens_per_minute": None,
53
+ },
54
+ "command-a": {
55
+ "id": "command-a",
56
+ "name": "command-a-03-2025",
57
+ "api_base": "https://api.cohere.ai/compatibility/v1",
58
+ "api_key_env_var": "COHERE_API_KEY",
59
+ "api_spec": "openai",
60
+ "input_cost": 0.5,
61
+ "output_cost": 1.5,
62
+ "requests_per_minute": 10_000,
63
+ "tokens_per_minute": None,
64
+ },
65
+ "command-r-7b": {
66
+ "id": "command-r-cohere",
67
+ "name": "command-r7b-12-2024",
68
+ "api_base": "https://api.cohere.ai/compatibility/v1",
69
+ "api_key_env_var": "COHERE_API_KEY",
70
+ "api_spec": "openai",
71
+ "input_cost": 0.5,
72
+ "output_cost": 1.5,
73
+ "requests_per_minute": 10_000,
74
+ "tokens_per_minute": None,
75
+ },
76
+ "command-r": {
77
+ "id": "command-r",
78
+ "name": "command-r-08-2024",
79
+ "api_base": "https://api.cohere.ai/compatibility/v1",
80
+ "api_key_env_var": "COHERE_API_KEY",
81
+ "api_spec": "openai",
82
+ "input_cost": 0.5,
83
+ "output_cost": 1.5,
84
+ "requests_per_minute": 10_000,
85
+ "tokens_per_minute": None,
86
+ },
87
+ "command-r-plus": {
88
+ "id": "command-r-plus",
89
+ "name": "command-r-plus-04-2024",
90
+ "api_base": "https://api.cohere.ai/compatibility/v1",
91
+ "api_key_env_var": "COHERE_API_KEY",
92
+ "api_spec": "openai",
93
+ "input_cost": 3.0,
94
+ "output_cost": 15.0,
95
+ "requests_per_minute": 10_000,
96
+ "tokens_per_minute": None,
97
+ },
98
+ }
lm_deluge/models/deepseek.py ADDED
@@ -0,0 +1,27 @@
1
+ DEEPSEEK_MODELS = {
2
+ # ______ _
3
+ # (______) | |
4
+ # _ _ _____ _____ ____ ___ _____ _____| | _
5
+ # | | | | ___ | ___ | _ \ /___) ___ | ___ | |_/ )
6
+ # | |__/ /| ____| ____| |_| |___ | ____| ____| _ (
7
+ # |_____/ |_____)_____) __/(___/|_____)_____)_| \_)
8
+ # |_|
9
+ "deepseek-chat": {
10
+ "id": "deepseek-chat",
11
+ "name": "deepseek-chat",
12
+ "api_base": "https://api.deepseek.com/v1",
13
+ "api_key_env_var": "DEEPSEEK_API_KEY",
14
+ "api_spec": "openai",
15
+ "input_cost": 0.27,
16
+ "output_cost": 1.10,
17
+ },
18
+ "deepseek-r1": {
19
+ "id": "deepseek-r1",
20
+ "name": "deepseek-reasoner",
21
+ "api_base": "https://api.deepseek.com/v1",
22
+ "api_key_env_var": "DEEPSEEK_API_KEY",
23
+ "api_spec": "openai",
24
+ "input_cost": 0.55,
25
+ "output_cost": 2.19,
26
+ },
27
+ }
lm_deluge/models/fireworks.py ADDED
@@ -0,0 +1,16 @@
1
+ FIREWORKS_MODELS = {
2
+ "gpt-oss-20b-fireworks": {
3
+ "id": "gpt-oss-20b-fireworks",
4
+ "name": "accounts/fireworks/models/gpt-oss-20b",
5
+ "api_base": "https://api.fireworks.ai/inference/v1",
6
+ "api_key_env_var": "FIREWORKS_API_KEY",
7
+ "api_spec": "openai",
8
+ },
9
+ "gpt-oss-120b-fireworks": {
10
+ "id": "gpt-oss-120b-fireworks",
11
+ "name": "accounts/fireworks/models/gpt-oss-120b",
12
+ "api_base": "https://api.fireworks.ai/inference/v1",
13
+ "api_key_env_var": "FIREWORKS_API_KEY",
14
+ "api_spec": "openai",
15
+ },
16
+ }
lm_deluge/models/google.py ADDED
@@ -0,0 +1,153 @@
1
+ GOOGLE_MODELS = {
2
+ # .oooooo. oooo .o. ooooo
3
+ # d8P' `Y8b `888 .888. `888'
4
+ # 888 .ooooo. .ooooo. .oooooooo 888 .ooooo. .8"888. 888
5
+ # 888 d88' `88b d88' `88b 888' `88b 888 d88' `88b .8' `888. 888
6
+ # 888 ooooo 888 888 888 888 888 888 888 888ooo888 .88ooo8888. 888
7
+ # `88. .88' 888 888 888 888 `88bod8P' 888 888 .o .8' `888. 888
8
+ # `Y8bood8P' `Y8bod8P' `Y8bod8P' `8oooooo. o888o `Y8bod8P' o88o o8888o o888o
9
+ # d" YD
10
+ # "Y88888P'
11
+ # these are through AI studio rather than Vertex, and using the OpenAI-compatible endpoints
12
+ "gemini-2.0-flash-compat": {
13
+ "id": "gemini-2.0-flash-compat",
14
+ "name": "gemini-2.0-flash",
15
+ "api_base": "https://generativelanguage.googleapis.com/v1beta/openai",
16
+ "api_key_env_var": "GEMINI_API_KEY",
17
+ "supports_json": True,
18
+ "supports_logprobs": False,
19
+ "api_spec": "openai",
20
+ "input_cost": 0.1,
21
+ "output_cost": 0.4,
22
+ "requests_per_minute": 20,
23
+ "tokens_per_minute": 100_000,
24
+ "reasoning_model": False,
25
+ },
26
+ "gemini-2.0-flash-lite-compat": {
27
+ "id": "gemini-2.0-flash-lite-compat",
28
+ "name": "gemini-2.0-flash-lite",
29
+ "api_base": "https://generativelanguage.googleapis.com/v1beta/openai",
30
+ "api_key_env_var": "GEMINI_API_KEY",
31
+ "supports_json": True,
32
+ "supports_logprobs": False,
33
+ "api_spec": "openai",
34
+ "input_cost": 0.1,
35
+ "output_cost": 0.4,
36
+ "requests_per_minute": 20,
37
+ "tokens_per_minute": 100_000,
38
+ "reasoning_model": False,
39
+ },
40
+ "gemini-2.5-pro-compat": {
41
+ "id": "gemini-2.5-pro-compat",
42
+ "name": "gemini-2.5-pro",
43
+ "api_base": "https://generativelanguage.googleapis.com/v1beta/openai",
44
+ "api_key_env_var": "GEMINI_API_KEY",
45
+ "supports_json": True,
46
+ "supports_logprobs": False,
47
+ "api_spec": "openai",
48
+ "input_cost": 0.1,
49
+ "output_cost": 0.4,
50
+ "requests_per_minute": 20,
51
+ "tokens_per_minute": 100_000,
52
+ "reasoning_model": True,
53
+ },
54
+ "gemini-2.5-flash-compat": {
55
+ "id": "gemini-2.5-flash-compat",
56
+ "name": "gemini-2.5-flash",
57
+ "api_base": "https://generativelanguage.googleapis.com/v1beta/openai",
58
+ "api_key_env_var": "GEMINI_API_KEY",
59
+ "supports_json": True,
60
+ "supports_logprobs": False,
61
+ "api_spec": "openai",
62
+ "input_cost": 0.1,
63
+ "output_cost": 0.4,
64
+ "requests_per_minute": 20,
65
+ "tokens_per_minute": 100_000,
66
+ "reasoning_model": True,
67
+ },
68
+ "gemini-2.5-flash-lite-compat": {
69
+ "id": "gemini-2.5-flash-lite-compat",
70
+ "name": "gemini-2.5-flash-lite",
71
+ "api_base": "https://generativelanguage.googleapis.com/v1beta/openai",
72
+ "api_key_env_var": "GEMINI_API_KEY",
73
+ "supports_json": True,
74
+ "supports_logprobs": False,
75
+ "api_spec": "openai",
76
+ "input_cost": 0.1,
77
+ "output_cost": 0.4,
78
+ "requests_per_minute": 20,
79
+ "tokens_per_minute": 100_000,
80
+ "reasoning_model": True,
81
+ },
82
+ # Native Gemini API versions with file support
83
+ "gemini-2.0-flash": {
84
+ "id": "gemini-2.0-flash",
85
+ "name": "gemini-2.0-flash",
86
+ "api_base": "https://generativelanguage.googleapis.com/v1beta",
87
+ "api_key_env_var": "GEMINI_API_KEY",
88
+ "supports_json": True,
89
+ "supports_logprobs": False,
90
+ "api_spec": "gemini",
91
+ "input_cost": 0.1,
92
+ "output_cost": 0.4,
93
+ "requests_per_minute": 20,
94
+ "tokens_per_minute": 100_000,
95
+ "reasoning_model": False,
96
+ },
97
+ "gemini-2.0-flash-lite": {
98
+ "id": "gemini-2.0-flash-lite",
99
+ "name": "gemini-2.0-flash-lite",
100
+ "api_base": "https://generativelanguage.googleapis.com/v1beta",
101
+ "api_key_env_var": "GEMINI_API_KEY",
102
+ "supports_json": True,
103
+ "supports_logprobs": False,
104
+ "api_spec": "gemini",
105
+ "input_cost": 0.1,
106
+ "output_cost": 0.4,
107
+ "requests_per_minute": 20,
108
+ "tokens_per_minute": 100_000,
109
+ "reasoning_model": False,
110
+ },
111
+ "gemini-2.5-pro": {
112
+ "id": "gemini-2.5-pro",
113
+ "name": "gemini-2.5-pro",
114
+ "api_base": "https://generativelanguage.googleapis.com/v1beta",
115
+ "api_key_env_var": "GEMINI_API_KEY",
116
+ "supports_json": True,
117
+ "supports_logprobs": False,
118
+ "api_spec": "gemini",
119
+ "input_cost": 0.1,
120
+ "output_cost": 0.4,
121
+ "requests_per_minute": 20,
122
+ "tokens_per_minute": 100_000,
123
+ "reasoning_model": True,
124
+ },
125
+ "gemini-2.5-flash": {
126
+ "id": "gemini-2.5-flash",
127
+ "name": "gemini-2.5-flash",
128
+ "api_base": "https://generativelanguage.googleapis.com/v1beta",
129
+ "api_key_env_var": "GEMINI_API_KEY",
130
+ "supports_json": True,
131
+ "supports_logprobs": False,
132
+ "api_spec": "gemini",
133
+ "input_cost": 0.1,
134
+ "output_cost": 0.4,
135
+ "requests_per_minute": 20,
136
+ "tokens_per_minute": 100_000,
137
+ "reasoning_model": True,
138
+ },
139
+ "gemini-2.5-flash-lite": {
140
+ "id": "gemini-2.5-flash-lite",
141
+ "name": "gemini-2.5-flash-lite",
142
+ "api_base": "https://generativelanguage.googleapis.com/v1beta",
143
+ "api_key_env_var": "GEMINI_API_KEY",
144
+ "supports_json": True,
145
+ "supports_logprobs": False,
146
+ "api_spec": "gemini",
147
+ "input_cost": 0.1,
148
+ "output_cost": 0.4,
149
+ "requests_per_minute": 20,
150
+ "tokens_per_minute": 100_000,
151
+ "reasoning_model": True,
152
+ },
153
+ }
lm_deluge/models/grok.py ADDED
@@ -0,0 +1,38 @@
1
+ XAI_MODELS = {
2
+ # .d8888b. 888
3
+ # d88P Y88b 888
4
+ # 888 888 888
5
+ # 888 888d888 .d88b. 888 888
6
+ # 888 88888 888P" d88""88b 888 .88P
7
+ # 888 888 888 888 888 888888K
8
+ # Y88b d88P 888 Y88..88P 888 "88b
9
+ # "Y8888P88 888 "Y88P" 888 888
10
+ "grok-3": {
11
+ "id": "grok-3",
12
+ "name": "grok-3-latest",
13
+ "api_base": "https://api.x.ai/v1",
14
+ "api_key_env_var": "GROK_API_KEY",
15
+ "supports_json": True,
16
+ "supports_logprobs": True,
17
+ "api_spec": "openai",
18
+ "input_cost": 2.0,
19
+ "output_cost": 8.0,
20
+ "requests_per_minute": 20,
21
+ "tokens_per_minute": 100_000,
22
+ "reasoning_model": False,
23
+ },
24
+ "grok-3-mini": {
25
+ "id": "grok-3-mini",
26
+ "name": "grok-3-mini-latest",
27
+ "api_base": "https://api.x.ai/v1",
28
+ "api_key_env_var": "GROK_API_KEY",
29
+ "supports_json": True,
30
+ "supports_logprobs": True,
31
+ "api_spec": "openai",
32
+ "input_cost": 2.0,
33
+ "output_cost": 8.0,
34
+ "requests_per_minute": 20,
35
+ "tokens_per_minute": 100_000,
36
+ "reasoning_model": True,
37
+ },
38
+ }