diff --git a/litellm/constants.py b/litellm/constants.py index 97631b9997a..3e4febb58fb 100644 --- a/litellm/constants.py +++ b/litellm/constants.py @@ -1092,6 +1092,7 @@ openai_compatible_endpoints: Final[list[str]] = [ "https://api.scx.ai/v1", "https://api.prisminference.com/v1", "https://api.reka.ai/v1", + "https://api.iteracompute.com/v1", "https://gigachat.devices.sberbank.ru/api/v1", "https://api.openinfer.ai/v1", "https://vanchin.streamlake.ai/api/gateway/v1/endpoints", @@ -1179,6 +1180,7 @@ openai_compatible_providers: Final[list[str]] = [ "sail", "reka", "streamlake", + "iteracompute", ] OPENAI_AUDIO_TRANSCRIPTION_PROVIDERS: Final = frozenset({"openai"} | frozenset(openai_compatible_providers)) diff --git a/litellm/llms/openai_like/providers.json b/litellm/llms/openai_like/providers.json index a71a9c3bb9b..0846b217d18 100644 --- a/litellm/llms/openai_like/providers.json +++ b/litellm/llms/openai_like/providers.json @@ -262,6 +262,11 @@ }, "supported_endpoints": ["/v1/chat/completions", "/v1/responses", "/v1/messages"] }, + "iteracompute": { + "base_url": "https://api.iteracompute.com/v1", + "api_key_env": "ITERACOMPUTE_API_KEY", + "api_base_env": "ITERACOMPUTE_API_BASE" + }, "inferenceapis": { "base_url": "https://api.inferenceapis.com/v1", "api_key_env": "INFERENCEAPIS_API_KEY", diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index be93d6fcd9b..67dd4da0177 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -34847,6 +34847,521 @@ "supports_video_input": true, "supports_vision": true }, + "iteracompute/anthropic/claude-fable-5-1": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 5e-06, + "output_cost_per_token_batches": 1.25e-05, + "cache_creation_input_token_cost": 6.25e-06, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 1e-05, + "output_cost_per_token": 2.5e-05, + "cache_read_input_token_cost": 1.25e-07, + "input_cost_per_token_batches": 2.5e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-haiku-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "tiered_pricing": [ + { + "range": [ + 0, + 100000 + ], + "input_cost_per_token": 5e-08, + "output_cost_per_token": 2.5e-07, + "cache_read_input_token_cost": 5e-09, + "cache_creation_input_token_cost": 6.25e-08, + "cache_creation_input_token_cost_above_1hr": 1e-07 + }, + { + "range": [ + 100000, + 1000000 + ], + "input_cost_per_token": 2.5e-07, + "output_cost_per_token": 1.25e-06, + "cache_read_input_token_cost": 2.5e-08, + "cache_creation_input_token_cost": 3.125e-07, + "cache_creation_input_token_cost_above_1hr": 5e-07 + } + ], + "input_cost_per_token": 5e-08, + "output_cost_per_token_batches": 1.25e-07, + "cache_creation_input_token_cost": 6.25e-08, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 1e-07, + "output_cost_per_token": 2.5e-07, + "cache_read_input_token_cost": 5e-09, + "input_cost_per_token_batches": 2.5e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-opus-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 2e-06, + "output_cost_per_token_batches": 5e-06, + "cache_creation_input_token_cost": 2.5e-06, + "provider_specific_entry": { + "us": 1.1, + "fast": 2 + }, + "cache_creation_input_token_cost_above_1hr": 4e-06, + "output_cost_per_token": 1e-05, + "cache_read_input_token_cost": 1e-07, + "input_cost_per_token_batches": 1e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-sonnet-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 1e-06, + "output_cost_per_token_batches": 2.5e-06, + "cache_creation_input_token_cost": 1.25e-06, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 2e-06, + "output_cost_per_token": 5e-06, + "cache_read_input_token_cost": 5e-08, + "input_cost_per_token_batches": 5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/deepseek/deepseek-v4-pro-0813": { + "input_cost_per_token": 6e-07, + "off_peak_pricing": { + "windows": [ + { + "weekdays": [ + 1, + 2, + 3, + 4, + 5 + ], + "hours_utc": [ + "00:00-01:00", + "04:00-06:00", + "10:00-00:00" + ] + }, + { + "weekdays": [ + 6, + 7 + ], + "hours_utc": "00:00-00:00" + } + ], + "input_cost_per_token": 5.012658227848101e-07, + "output_cost_per_token": 1.4891596638655463e-06, + "cache_read_input_token_cost": 3.7230769230769236e-08 + }, + "output_cost_per_token": 1.79e-06, + "cache_read_input_token_cost": 4.4e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/deepseek/deepseek-v4.1-flash": { + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 5.5e-07, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/deepseek/deepseek-v4.1-flash-1017": { + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 5.5e-07, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/gemini/gemini-3.8-flash": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 3e-08, + "google_maps_grounding_cost_per_query": 0.0112, + "output_cost_per_reasoning_token": 3e-06, + "input_cost_per_token": 6e-07, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 3e-07, + "output_cost_per_token_batches": 1.5e-06, + "cache_read_input_token_cost_priority": 1.08e-07, + "output_cost_per_token_flex": 1.5e-06, + "output_cost_per_token_priority": 5.4e-06, + "output_cost_per_reasoning_token_priority": 5.4e-06, + "output_cost_per_reasoning_token_flex": 1.5e-06, + "output_cost_per_token": 3e-06, + "cache_read_input_token_cost": 6e-08, + "input_cost_per_token_priority": 1.08e-06, + "input_cost_per_token_batches": 3e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 65536 + }, + "iteracompute/gemini/gemini-4-argon": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 4e-08, + "google_maps_grounding_cost_per_query": 0.0112, + "output_cost_per_reasoning_token": 8e-06, + "input_cost_per_token": 1.6e-06, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 8e-07, + "cache_read_input_token_cost_priority": 1.4400000000000002e-07, + "output_cost_per_token_flex": 4e-06, + "output_cost_per_token_priority": 1.44e-05, + "output_cost_per_reasoning_token_priority": 1.44e-05, + "output_cost_per_reasoning_token_flex": 4e-06, + "output_cost_per_token": 8e-06, + "cache_read_input_token_cost": 8e-08, + "input_cost_per_token_priority": 2.88e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 65536 + }, + "iteracompute/gemini/gemini-nano-banana-2.1": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 6e-08, + "output_cost_per_reasoning_token": 6e-06, + "input_cost_per_token": 1.2e-06, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 6e-07, + "output_cost_per_token_batches": 3e-06, + "cache_read_input_token_cost_priority": 2.16e-07, + "output_cost_per_token_flex": 3e-06, + "output_cost_per_token_priority": 1.08e-05, + "input_cost_per_video_token": 1.2e-06, + "input_cost_per_image_token": 1.2e-06, + "input_cost_per_image": 0.001344, + "output_cost_per_token": 6e-06, + "output_cost_per_image_token": 2.4e-05, + "output_cost_per_image": 0.02688, + "cache_read_input_token_cost": 1.2e-07, + "input_cost_per_token_priority": 2.16e-06, + "input_cost_per_token_batches": 6e-07, + "litellm_provider": "iteracompute", + "mode": "image_generation", + "max_input_tokens": 131072, + "max_output_tokens": 32768 + }, + "iteracompute/minimax/minimax-m2.5": { + "input_cost_per_token": 2.7e-07, + "output_cost_per_token": 1.03e-06, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 204800, + "max_output_tokens": 16384 + }, + "iteracompute/minimax/minimax-m2.7": { + "input_cost_per_token": 3.8e-07, + "output_cost_per_token": 1.48e-06, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 204800, + "max_output_tokens": 16384 + }, + "iteracompute/minimax/minimax-m3": { + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.19e-06, + "cache_read_input_token_cost": 5e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 524288 + }, + "iteracompute/moonshotai/kimi-k2.5": { + "input_cost_per_token": 5.4e-07, + "output_cost_per_token": 2.78e-06, + "cache_read_input_token_cost": 6e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 16384 + }, + "iteracompute/moonshotai/kimi-k2.6": { + "input_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.51e-06, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 16384 + }, + "iteracompute/moonshotai/kimi-k2.7-code": { + "input_cost_per_token": 6.9e-07, + "output_cost_per_token": 3.29e-06, + "cache_read_input_token_cost": 1.5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 262144 + }, + "iteracompute/moonshotai/kimi-k3": { + "input_cost_per_token": 2.35e-06, + "output_cost_per_token": 1.176e-05, + "cache_read_input_token_cost": 2.3e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 1048576 + }, + "iteracompute/openai/gpt-6-astra": { + "cache_read_input_token_cost_ultrafast": 4.8e-06, + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-07, + "input_cost_per_token_ultrafast": 4.8e-05, + "output_cost_per_token_above_272k_tokens_priority": 0.00012, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-06, + "input_cost_per_token": 8e-06, + "input_cost_per_token_flex": 4e-06, + "output_cost_per_token_batches": 2e-05, + "cache_creation_input_token_cost_above_272k_tokens_flex": 1e-05, + "cache_read_input_token_cost_priority": 1.6e-06, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 1e-05, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-07, + "output_cost_per_token_flex": 2e-05, + "cache_creation_input_token_cost_ultrafast": 6e-05, + "output_cost_per_token_above_272k_tokens": 6e-05, + "output_cost_per_token_above_272k_tokens_ultrafast": 0.00036, + "input_cost_per_token_above_272k_tokens_ultrafast": 9.6e-05, + "input_cost_per_token_above_272k_tokens_flex": 8e-06, + "output_cost_per_token_ultrafast": 0.00024, + "output_cost_per_token_priority": 8e-05, + "input_cost_per_token_above_272k_tokens_priority": 3.2e-05, + "cache_creation_input_token_cost_above_272k_tokens_priority": 4e-05, + "output_cost_per_token": 4e-05, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-06, + "cache_creation_input_token_cost_above_272k_tokens_ultrafast": 0.00012, + "cache_creation_input_token_cost_flex": 5e-06, + "cache_creation_input_token_cost_priority": 2e-05, + "cache_read_input_token_cost_above_272k_tokens_ultrafast": 9.6e-06, + "output_cost_per_token_above_272k_tokens_flex": 3e-05, + "cache_read_input_token_cost": 8e-07, + "input_cost_per_token_priority": 1.6e-05, + "input_cost_per_token_batches": 4e-06, + "cache_creation_input_token_cost_above_272k_tokens": 2e-05, + "input_cost_per_token_above_272k_tokens": 1.6e-05, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-6-luna": { + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-09, + "output_cost_per_token_above_272k_tokens_priority": 1.2e-06, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-08, + "input_cost_per_token": 8e-08, + "input_cost_per_token_flex": 4e-08, + "output_cost_per_token_batches": 2e-07, + "cache_creation_input_token_cost_above_272k_tokens_flex": 1e-07, + "cache_read_input_token_cost_priority": 1.6e-08, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 1e-07, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-09, + "output_cost_per_token_flex": 2e-07, + "output_cost_per_token_above_272k_tokens": 6e-07, + "input_cost_per_token_above_272k_tokens_flex": 8e-08, + "output_cost_per_token_priority": 8e-07, + "input_cost_per_token_above_272k_tokens_priority": 3.2e-07, + "cache_creation_input_token_cost_above_272k_tokens_priority": 4e-07, + "output_cost_per_token": 4e-07, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-08, + "cache_creation_input_token_cost_flex": 5e-08, + "cache_creation_input_token_cost_priority": 2e-07, + "output_cost_per_token_above_272k_tokens_flex": 3e-07, + "cache_read_input_token_cost": 8e-09, + "input_cost_per_token_priority": 1.6e-07, + "input_cost_per_token_batches": 4e-08, + "cache_creation_input_token_cost_above_272k_tokens": 2e-07, + "input_cost_per_token_above_272k_tokens": 1.6e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-6.1-sol": { + "cache_read_input_token_cost_ultrafast": 4.8e-07, + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-08, + "input_cost_per_token_ultrafast": 9.6e-06, + "output_cost_per_token_above_272k_tokens_priority": 2.4e-05, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-07, + "input_cost_per_token": 1.6e-06, + "input_cost_per_token_flex": 8e-07, + "output_cost_per_token_batches": 4e-06, + "cache_creation_input_token_cost_above_272k_tokens_flex": 2e-06, + "cache_read_input_token_cost_priority": 1.6e-07, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 2e-06, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-08, + "output_cost_per_token_flex": 4e-06, + "cache_creation_input_token_cost_ultrafast": 1.2e-05, + "output_cost_per_token_above_272k_tokens": 1.2e-05, + "output_cost_per_token_above_272k_tokens_ultrafast": 7.2e-05, + "input_cost_per_token_above_272k_tokens_ultrafast": 1.92e-05, + "input_cost_per_token_above_272k_tokens_flex": 1.6e-06, + "output_cost_per_token_ultrafast": 4.8e-05, + "output_cost_per_token_priority": 1.6e-05, + "input_cost_per_token_above_272k_tokens_priority": 6.4e-06, + "cache_creation_input_token_cost_above_272k_tokens_priority": 8e-06, + "output_cost_per_token": 8e-06, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-07, + "cache_creation_input_token_cost_above_272k_tokens_ultrafast": 2.4e-05, + "cache_creation_input_token_cost_flex": 1e-06, + "cache_creation_input_token_cost_priority": 4e-06, + "cache_read_input_token_cost_above_272k_tokens_ultrafast": 9.6e-07, + "output_cost_per_token_above_272k_tokens_flex": 6e-06, + "cache_read_input_token_cost": 8e-08, + "input_cost_per_token_priority": 3.2e-06, + "input_cost_per_token_batches": 8e-07, + "cache_creation_input_token_cost_above_272k_tokens": 4e-06, + "input_cost_per_token_above_272k_tokens": 3.2e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-image-2.5-flare": { + "input_cost_per_token": 4e-06, + "input_cost_per_image_token": 6.4e-06, + "output_cost_per_token": 0.0, + "output_cost_per_image_token": 2.4e-05, + "cache_read_input_token_cost": 1e-06, + "input_cost_per_token_batches": 2e-06, + "litellm_provider": "iteracompute", + "mode": "image_generation" + }, + "iteracompute/openai/gpt-image-2.5-sunburst": { + "input_cost_per_token": 4e-06, + "input_cost_per_image_token": 6.4e-06, + "output_cost_per_token": 0.0, + "output_cost_per_image_token": 2.4e-05, + "cache_read_input_token_cost": 1e-06, + "input_cost_per_token_batches": 2e-06, + "litellm_provider": "iteracompute", + "mode": "image_generation" + }, + "iteracompute/qwen/qwen3.8-27b": { + "input_cost_per_token": 0.0, + "output_cost_per_token": 0.0, + "cache_read_input_token_cost": 0.0, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 8192 + }, + "iteracompute/qwen/qwen3.8-flash": { + "input_cost_per_token": 1.1e-07, + "output_cost_per_token": 3.8e-07, + "cache_read_input_token_cost": 1.6e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/qwen/qwen3.8-max": { + "input_cost_per_token": 1.95e-06, + "output_cost_per_token": 5.7e-06, + "cache_read_input_token_cost": 2.5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.2": { + "input_cost_per_token": 5.3e-07, + "output_cost_per_token": 1.72e-06, + "cache_read_input_token_cost": 1.2e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.3": { + "input_cost_per_token": 8.2e-07, + "output_cost_per_token": 2.88e-06, + "cache_read_input_token_cost": 2e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.3-flash": { + "input_cost_per_token": 8e-08, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 2.3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, "gpt-3.5-turbo": { "deprecation_date": "2026-10-23", "input_cost_per_token": 5e-07, diff --git a/litellm/provider_endpoints_support_backup.json b/litellm/provider_endpoints_support_backup.json index 40ea9d15ff4..8ad96c24c24 100644 --- a/litellm/provider_endpoints_support_backup.json +++ b/litellm/provider_endpoints_support_backup.json @@ -33,6 +33,36 @@ } }, "providers": { + "iteracompute": { + "display_name": "IteraCompute (`iteracompute`)", + "url": "https://docs.litellm.ai/docs/providers/iteracompute", + "endpoints": { + "chat_completions": true, + "systemone": false, + "messages": true, + "responses": true, + "embeddings": false, + "image_generations": false, + "audio_transcriptions": false, + "audio_speech": false, + "moderations": false, + "batches": false, + "rerank": false, + "ocr": false, + "search": false, + "skills": false, + "interactions": false, + "a2a_(Agent Gateway)": false, + "container": false, + "container_file": false, + "compact": false, + "files": false, + "image_edits": false, + "vector_stores_create": false, + "vector_stores_search": false, + "video_generations": false + } + }, "refuseless": { "display_name": "Refuseless (`refuseless`)", "url": "https://docs.litellm.ai/docs/providers/refuseless", diff --git a/litellm/proxy/public_endpoints/provider_create_fields.json b/litellm/proxy/public_endpoints/provider_create_fields.json index aa7554c4a5a..641bed41501 100644 --- a/litellm/proxy/public_endpoints/provider_create_fields.json +++ b/litellm/proxy/public_endpoints/provider_create_fields.json @@ -4452,5 +4452,33 @@ } ], "default_model_placeholder": "bigmodel/glm-5.3" + }, + { + "provider": "ITERACOMPUTE", + "provider_display_name": "IteraCompute", + "litellm_provider": "iteracompute", + "credential_fields": [ + { + "key": "api_base", + "label": "API Base", + "placeholder": "https://api.iteracompute.com/v1", + "tooltip": null, + "required": false, + "field_type": "text", + "options": null, + "default_value": null + }, + { + "key": "api_key", + "label": "API Key", + "placeholder": null, + "tooltip": null, + "required": true, + "field_type": "password", + "options": null, + "default_value": null + } + ], + "default_model_placeholder": "iteracompute/qwen/qwen3.8-27b" } ] diff --git a/litellm/types/utils.py b/litellm/types/utils.py index 57f4d971eb3..1e1d8359b5c 100644 --- a/litellm/types/utils.py +++ b/litellm/types/utils.py @@ -4756,6 +4756,7 @@ class LlmProviders(str, Enum): SAIL = "sail" REKA = "reka" STREAMLAKE = "streamlake" + ITERACOMPUTE = "iteracompute" LITELLM_AGENT = "litellm_agent" CURSOR = "cursor" BEDROCK_MANTLE = "bedrock_mantle" diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index be93d6fcd9b..67dd4da0177 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -34847,6 +34847,521 @@ "supports_video_input": true, "supports_vision": true }, + "iteracompute/anthropic/claude-fable-5-1": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 5e-06, + "output_cost_per_token_batches": 1.25e-05, + "cache_creation_input_token_cost": 6.25e-06, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 1e-05, + "output_cost_per_token": 2.5e-05, + "cache_read_input_token_cost": 1.25e-07, + "input_cost_per_token_batches": 2.5e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-haiku-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "tiered_pricing": [ + { + "range": [ + 0, + 100000 + ], + "input_cost_per_token": 5e-08, + "output_cost_per_token": 2.5e-07, + "cache_read_input_token_cost": 5e-09, + "cache_creation_input_token_cost": 6.25e-08, + "cache_creation_input_token_cost_above_1hr": 1e-07 + }, + { + "range": [ + 100000, + 1000000 + ], + "input_cost_per_token": 2.5e-07, + "output_cost_per_token": 1.25e-06, + "cache_read_input_token_cost": 2.5e-08, + "cache_creation_input_token_cost": 3.125e-07, + "cache_creation_input_token_cost_above_1hr": 5e-07 + } + ], + "input_cost_per_token": 5e-08, + "output_cost_per_token_batches": 1.25e-07, + "cache_creation_input_token_cost": 6.25e-08, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 1e-07, + "output_cost_per_token": 2.5e-07, + "cache_read_input_token_cost": 5e-09, + "input_cost_per_token_batches": 2.5e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-opus-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 2e-06, + "output_cost_per_token_batches": 5e-06, + "cache_creation_input_token_cost": 2.5e-06, + "provider_specific_entry": { + "us": 1.1, + "fast": 2 + }, + "cache_creation_input_token_cost_above_1hr": 4e-06, + "output_cost_per_token": 1e-05, + "cache_read_input_token_cost": 1e-07, + "input_cost_per_token_batches": 1e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/anthropic/claude-sonnet-5-5": { + "search_context_cost_per_query": { + "search_context_size_low": 0.005, + "search_context_size_high": 0.005, + "search_context_size_medium": 0.005 + }, + "input_cost_per_token": 1e-06, + "output_cost_per_token_batches": 2.5e-06, + "cache_creation_input_token_cost": 1.25e-06, + "provider_specific_entry": { + "us": 1.1 + }, + "cache_creation_input_token_cost_above_1hr": 2e-06, + "output_cost_per_token": 5e-06, + "cache_read_input_token_cost": 5e-08, + "input_cost_per_token_batches": 5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1000000, + "max_output_tokens": 128000 + }, + "iteracompute/deepseek/deepseek-v4-pro-0813": { + "input_cost_per_token": 6e-07, + "off_peak_pricing": { + "windows": [ + { + "weekdays": [ + 1, + 2, + 3, + 4, + 5 + ], + "hours_utc": [ + "00:00-01:00", + "04:00-06:00", + "10:00-00:00" + ] + }, + { + "weekdays": [ + 6, + 7 + ], + "hours_utc": "00:00-00:00" + } + ], + "input_cost_per_token": 5.012658227848101e-07, + "output_cost_per_token": 1.4891596638655463e-06, + "cache_read_input_token_cost": 3.7230769230769236e-08 + }, + "output_cost_per_token": 1.79e-06, + "cache_read_input_token_cost": 4.4e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/deepseek/deepseek-v4.1-flash": { + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 5.5e-07, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/deepseek/deepseek-v4.1-flash-1017": { + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 5.5e-07, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 393216 + }, + "iteracompute/gemini/gemini-3.8-flash": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 3e-08, + "google_maps_grounding_cost_per_query": 0.0112, + "output_cost_per_reasoning_token": 3e-06, + "input_cost_per_token": 6e-07, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 3e-07, + "output_cost_per_token_batches": 1.5e-06, + "cache_read_input_token_cost_priority": 1.08e-07, + "output_cost_per_token_flex": 1.5e-06, + "output_cost_per_token_priority": 5.4e-06, + "output_cost_per_reasoning_token_priority": 5.4e-06, + "output_cost_per_reasoning_token_flex": 1.5e-06, + "output_cost_per_token": 3e-06, + "cache_read_input_token_cost": 6e-08, + "input_cost_per_token_priority": 1.08e-06, + "input_cost_per_token_batches": 3e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 65536 + }, + "iteracompute/gemini/gemini-4-argon": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 4e-08, + "google_maps_grounding_cost_per_query": 0.0112, + "output_cost_per_reasoning_token": 8e-06, + "input_cost_per_token": 1.6e-06, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 8e-07, + "cache_read_input_token_cost_priority": 1.4400000000000002e-07, + "output_cost_per_token_flex": 4e-06, + "output_cost_per_token_priority": 1.44e-05, + "output_cost_per_reasoning_token_priority": 1.44e-05, + "output_cost_per_reasoning_token_flex": 4e-06, + "output_cost_per_token": 8e-06, + "cache_read_input_token_cost": 8e-08, + "input_cost_per_token_priority": 2.88e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 65536 + }, + "iteracompute/gemini/gemini-nano-banana-2.1": { + "search_context_cost_per_query": { + "search_context_size_low": 0.0112, + "search_context_size_high": 0.0112, + "search_context_size_medium": 0.0112 + }, + "cache_read_input_token_cost_flex": 6e-08, + "output_cost_per_reasoning_token": 6e-06, + "input_cost_per_token": 1.2e-06, + "web_search_billing_unit": "per_query", + "input_cost_per_token_flex": 6e-07, + "output_cost_per_token_batches": 3e-06, + "cache_read_input_token_cost_priority": 2.16e-07, + "output_cost_per_token_flex": 3e-06, + "output_cost_per_token_priority": 1.08e-05, + "input_cost_per_video_token": 1.2e-06, + "input_cost_per_image_token": 1.2e-06, + "input_cost_per_image": 0.001344, + "output_cost_per_token": 6e-06, + "output_cost_per_image_token": 2.4e-05, + "output_cost_per_image": 0.02688, + "cache_read_input_token_cost": 1.2e-07, + "input_cost_per_token_priority": 2.16e-06, + "input_cost_per_token_batches": 6e-07, + "litellm_provider": "iteracompute", + "mode": "image_generation", + "max_input_tokens": 131072, + "max_output_tokens": 32768 + }, + "iteracompute/minimax/minimax-m2.5": { + "input_cost_per_token": 2.7e-07, + "output_cost_per_token": 1.03e-06, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 204800, + "max_output_tokens": 16384 + }, + "iteracompute/minimax/minimax-m2.7": { + "input_cost_per_token": 3.8e-07, + "output_cost_per_token": 1.48e-06, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 204800, + "max_output_tokens": 16384 + }, + "iteracompute/minimax/minimax-m3": { + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.19e-06, + "cache_read_input_token_cost": 5e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 524288 + }, + "iteracompute/moonshotai/kimi-k2.5": { + "input_cost_per_token": 5.4e-07, + "output_cost_per_token": 2.78e-06, + "cache_read_input_token_cost": 6e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 16384 + }, + "iteracompute/moonshotai/kimi-k2.6": { + "input_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.51e-06, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 16384 + }, + "iteracompute/moonshotai/kimi-k2.7-code": { + "input_cost_per_token": 6.9e-07, + "output_cost_per_token": 3.29e-06, + "cache_read_input_token_cost": 1.5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 262144 + }, + "iteracompute/moonshotai/kimi-k3": { + "input_cost_per_token": 2.35e-06, + "output_cost_per_token": 1.176e-05, + "cache_read_input_token_cost": 2.3e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 1048576 + }, + "iteracompute/openai/gpt-6-astra": { + "cache_read_input_token_cost_ultrafast": 4.8e-06, + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-07, + "input_cost_per_token_ultrafast": 4.8e-05, + "output_cost_per_token_above_272k_tokens_priority": 0.00012, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-06, + "input_cost_per_token": 8e-06, + "input_cost_per_token_flex": 4e-06, + "output_cost_per_token_batches": 2e-05, + "cache_creation_input_token_cost_above_272k_tokens_flex": 1e-05, + "cache_read_input_token_cost_priority": 1.6e-06, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 1e-05, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-07, + "output_cost_per_token_flex": 2e-05, + "cache_creation_input_token_cost_ultrafast": 6e-05, + "output_cost_per_token_above_272k_tokens": 6e-05, + "output_cost_per_token_above_272k_tokens_ultrafast": 0.00036, + "input_cost_per_token_above_272k_tokens_ultrafast": 9.6e-05, + "input_cost_per_token_above_272k_tokens_flex": 8e-06, + "output_cost_per_token_ultrafast": 0.00024, + "output_cost_per_token_priority": 8e-05, + "input_cost_per_token_above_272k_tokens_priority": 3.2e-05, + "cache_creation_input_token_cost_above_272k_tokens_priority": 4e-05, + "output_cost_per_token": 4e-05, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-06, + "cache_creation_input_token_cost_above_272k_tokens_ultrafast": 0.00012, + "cache_creation_input_token_cost_flex": 5e-06, + "cache_creation_input_token_cost_priority": 2e-05, + "cache_read_input_token_cost_above_272k_tokens_ultrafast": 9.6e-06, + "output_cost_per_token_above_272k_tokens_flex": 3e-05, + "cache_read_input_token_cost": 8e-07, + "input_cost_per_token_priority": 1.6e-05, + "input_cost_per_token_batches": 4e-06, + "cache_creation_input_token_cost_above_272k_tokens": 2e-05, + "input_cost_per_token_above_272k_tokens": 1.6e-05, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-6-luna": { + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-09, + "output_cost_per_token_above_272k_tokens_priority": 1.2e-06, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-08, + "input_cost_per_token": 8e-08, + "input_cost_per_token_flex": 4e-08, + "output_cost_per_token_batches": 2e-07, + "cache_creation_input_token_cost_above_272k_tokens_flex": 1e-07, + "cache_read_input_token_cost_priority": 1.6e-08, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 1e-07, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-09, + "output_cost_per_token_flex": 2e-07, + "output_cost_per_token_above_272k_tokens": 6e-07, + "input_cost_per_token_above_272k_tokens_flex": 8e-08, + "output_cost_per_token_priority": 8e-07, + "input_cost_per_token_above_272k_tokens_priority": 3.2e-07, + "cache_creation_input_token_cost_above_272k_tokens_priority": 4e-07, + "output_cost_per_token": 4e-07, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-08, + "cache_creation_input_token_cost_flex": 5e-08, + "cache_creation_input_token_cost_priority": 2e-07, + "output_cost_per_token_above_272k_tokens_flex": 3e-07, + "cache_read_input_token_cost": 8e-09, + "input_cost_per_token_priority": 1.6e-07, + "input_cost_per_token_batches": 4e-08, + "cache_creation_input_token_cost_above_272k_tokens": 2e-07, + "input_cost_per_token_above_272k_tokens": 1.6e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-6.1-sol": { + "cache_read_input_token_cost_ultrafast": 4.8e-07, + "search_context_cost_per_query": { + "search_context_size_low": 0.008, + "search_context_size_high": 0.008, + "search_context_size_medium": 0.008 + }, + "cache_read_input_token_cost_flex": 4e-08, + "input_cost_per_token_ultrafast": 9.6e-06, + "output_cost_per_token_above_272k_tokens_priority": 2.4e-05, + "cache_read_input_token_cost_above_272k_tokens_priority": 3.2e-07, + "input_cost_per_token": 1.6e-06, + "input_cost_per_token_flex": 8e-07, + "output_cost_per_token_batches": 4e-06, + "cache_creation_input_token_cost_above_272k_tokens_flex": 2e-06, + "cache_read_input_token_cost_priority": 1.6e-07, + "regional_processing_uplift_multiplier_us": 1.1, + "cache_creation_input_token_cost": 2e-06, + "regional_processing_uplift_multiplier_eu": 1.1, + "cache_read_input_token_cost_above_272k_tokens_flex": 8e-08, + "output_cost_per_token_flex": 4e-06, + "cache_creation_input_token_cost_ultrafast": 1.2e-05, + "output_cost_per_token_above_272k_tokens": 1.2e-05, + "output_cost_per_token_above_272k_tokens_ultrafast": 7.2e-05, + "input_cost_per_token_above_272k_tokens_ultrafast": 1.92e-05, + "input_cost_per_token_above_272k_tokens_flex": 1.6e-06, + "output_cost_per_token_ultrafast": 4.8e-05, + "output_cost_per_token_priority": 1.6e-05, + "input_cost_per_token_above_272k_tokens_priority": 6.4e-06, + "cache_creation_input_token_cost_above_272k_tokens_priority": 8e-06, + "output_cost_per_token": 8e-06, + "cache_read_input_token_cost_above_272k_tokens": 1.6e-07, + "cache_creation_input_token_cost_above_272k_tokens_ultrafast": 2.4e-05, + "cache_creation_input_token_cost_flex": 1e-06, + "cache_creation_input_token_cost_priority": 4e-06, + "cache_read_input_token_cost_above_272k_tokens_ultrafast": 9.6e-07, + "output_cost_per_token_above_272k_tokens_flex": 6e-06, + "cache_read_input_token_cost": 8e-08, + "input_cost_per_token_priority": 3.2e-06, + "input_cost_per_token_batches": 8e-07, + "cache_creation_input_token_cost_above_272k_tokens": 4e-06, + "input_cost_per_token_above_272k_tokens": 3.2e-06, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1050000, + "max_output_tokens": 128000 + }, + "iteracompute/openai/gpt-image-2.5-flare": { + "input_cost_per_token": 4e-06, + "input_cost_per_image_token": 6.4e-06, + "output_cost_per_token": 0.0, + "output_cost_per_image_token": 2.4e-05, + "cache_read_input_token_cost": 1e-06, + "input_cost_per_token_batches": 2e-06, + "litellm_provider": "iteracompute", + "mode": "image_generation" + }, + "iteracompute/openai/gpt-image-2.5-sunburst": { + "input_cost_per_token": 4e-06, + "input_cost_per_image_token": 6.4e-06, + "output_cost_per_token": 0.0, + "output_cost_per_image_token": 2.4e-05, + "cache_read_input_token_cost": 1e-06, + "input_cost_per_token_batches": 2e-06, + "litellm_provider": "iteracompute", + "mode": "image_generation" + }, + "iteracompute/qwen/qwen3.8-27b": { + "input_cost_per_token": 0.0, + "output_cost_per_token": 0.0, + "cache_read_input_token_cost": 0.0, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 262144, + "max_output_tokens": 8192 + }, + "iteracompute/qwen/qwen3.8-flash": { + "input_cost_per_token": 1.1e-07, + "output_cost_per_token": 3.8e-07, + "cache_read_input_token_cost": 1.6e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/qwen/qwen3.8-max": { + "input_cost_per_token": 1.95e-06, + "output_cost_per_token": 5.7e-06, + "cache_read_input_token_cost": 2.5e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.2": { + "input_cost_per_token": 5.3e-07, + "output_cost_per_token": 1.72e-06, + "cache_read_input_token_cost": 1.2e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.3": { + "input_cost_per_token": 8.2e-07, + "output_cost_per_token": 2.88e-06, + "cache_read_input_token_cost": 2e-07, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, + "iteracompute/z-ai/glm-5.3-flash": { + "input_cost_per_token": 8e-08, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 2.3e-08, + "litellm_provider": "iteracompute", + "mode": "chat", + "max_input_tokens": 1048576, + "max_output_tokens": 131072 + }, "gpt-3.5-turbo": { "deprecation_date": "2026-10-23", "input_cost_per_token": 5e-07, diff --git a/provider_endpoints_support.json b/provider_endpoints_support.json index 40ea9d15ff4..8ad96c24c24 100644 --- a/provider_endpoints_support.json +++ b/provider_endpoints_support.json @@ -33,6 +33,36 @@ } }, "providers": { + "iteracompute": { + "display_name": "IteraCompute (`iteracompute`)", + "url": "https://docs.litellm.ai/docs/providers/iteracompute", + "endpoints": { + "chat_completions": true, + "systemone": false, + "messages": true, + "responses": true, + "embeddings": false, + "image_generations": false, + "audio_transcriptions": false, + "audio_speech": false, + "moderations": false, + "batches": false, + "rerank": false, + "ocr": false, + "search": false, + "skills": false, + "interactions": false, + "a2a_(Agent Gateway)": false, + "container": false, + "container_file": false, + "compact": false, + "files": false, + "image_edits": false, + "vector_stores_create": false, + "vector_stores_search": false, + "video_generations": false + } + }, "refuseless": { "display_name": "Refuseless (`refuseless`)", "url": "https://docs.litellm.ai/docs/providers/refuseless", diff --git a/tests/unit/llms/openai_like/test_iteracompute_provider.py b/tests/unit/llms/openai_like/test_iteracompute_provider.py new file mode 100644 index 00000000000..039a432d628 --- /dev/null +++ b/tests/unit/llms/openai_like/test_iteracompute_provider.py @@ -0,0 +1,175 @@ +import json +from typing import Final + +import pytest +import respx + +import litellm +from litellm.caching.llm_caching_handler import LLMClientCache +from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider + +_COMPLETION: Final = { + "id": "chatcmpl_test", + "object": "chat.completion", + "created": 1_790_000_000, + "model": "qwen/qwen3.8-27b", + "choices": [ + { + "index": 0, + "message": {"role": "assistant", "content": "Hello"}, + "finish_reason": "stop", + } + ], + "usage": {"prompt_tokens": 4, "completion_tokens": 3, "total_tokens": 7}, +} + + +@pytest.mark.parametrize( + "model,custom_provider,api_base", + ( + ("iteracompute/qwen/qwen3.8-27b", None, None), + ("qwen/qwen3.8-27b", "iteracompute", None), + ("qwen/qwen3.8-27b", None, "https://api.iteracompute.com/v1"), + ), +) +def test_provider_resolution_preserves_model_namespace( + monkeypatch: pytest.MonkeyPatch, + model: str, + custom_provider: str | None, + api_base: str | None, +) -> None: + monkeypatch.setenv("ITERACOMPUTE_API_KEY", "test-key") + monkeypatch.delenv("ITERACOMPUTE_API_BASE", raising=False) + + assert get_llm_provider( + model=model, + custom_llm_provider=custom_provider, + api_base=api_base, + api_key=None, + ) == ( + "qwen/qwen3.8-27b", + "iteracompute", + "test-key", + "https://api.iteracompute.com/v1", + ) + + +@pytest.mark.parametrize("explicit", (False, True)) +def test_credentials_respect_explicit_arguments_before_environment( + monkeypatch: pytest.MonkeyPatch, explicit: bool +) -> None: + monkeypatch.setenv("ITERACOMPUTE_API_KEY", "env-key") + monkeypatch.setenv("ITERACOMPUTE_API_BASE", "https://env.example/v1") + + assert get_llm_provider( + model="iteracompute/qwen/qwen3.8-27b", + custom_llm_provider=None, + api_base="https://explicit.example/v1" if explicit else None, + api_key="explicit-key" if explicit else None, + ) == ( + "qwen/qwen3.8-27b", + "iteracompute", + "explicit-key" if explicit else "env-key", + "https://explicit.example/v1" if explicit else "https://env.example/v1", + ) + + +def test_completion_sends_model_and_bearer_key_without_provider_prefix() -> None: + with respx.mock() as upstream: + route: Final = upstream.post("https://api.iteracompute.com/v1/chat/completions").respond(200, json=_COMPLETION) + response: Final = litellm.completion( + model="iteracompute/qwen/qwen3.8-27b", + messages=[{"role": "user", "content": "Hello"}], + api_key="test-key", + max_tokens=512, + ) + + request: Final = route.calls.last.request + assert request.headers["authorization"] == "Bearer test-key" + assert json.loads(request.content) == { + "model": "qwen/qwen3.8-27b", + "messages": [{"role": "user", "content": "Hello"}], + "max_tokens": 512, + } + assert response.choices[0].message.content == "Hello" + assert response.usage.total_tokens == 7 + + +def test_streaming_returns_text_and_usage() -> None: + event: Final = { + "id": "chatcmpl_test", + "object": "chat.completion.chunk", + "created": 1_790_000_000, + "model": "qwen/qwen3.8-27b", + "choices": [ + { + "index": 0, + "delta": {"role": "assistant", "content": "Hello"}, + "finish_reason": "stop", + } + ], + "usage": {"prompt_tokens": 4, "completion_tokens": 3, "total_tokens": 7}, + } + with respx.mock() as upstream: + route: Final = upstream.post("https://api.iteracompute.com/v1/chat/completions").respond( + 200, + text=f"data: {json.dumps(event)}\n\ndata: [DONE]\n\n", + headers={"content-type": "text/event-stream"}, + ) + chunks: Final = tuple( + litellm.completion( + model="iteracompute/qwen/qwen3.8-27b", + messages=[{"role": "user", "content": "Hello"}], + api_key="test-key", + stream=True, + stream_options={"include_usage": True}, + ) + ) + + body: Final = json.loads(route.calls.last.request.content) + assert body["model"] == "qwen/qwen3.8-27b" + assert body["stream_options"] == {"include_usage": True} + assert "".join(chunk.choices[0].delta.content or "" for chunk in chunks if chunk.choices) == "Hello" + assert next(chunk.usage for chunk in chunks if getattr(chunk, "usage", None) is not None).total_tokens == 7 + + +def test_invalid_key_surfaces_authentication_error() -> None: + with respx.mock() as upstream: + upstream.post("https://api.iteracompute.com/v1/chat/completions").respond( + 401, + json={"error": {"message": "Invalid API key", "type": "authentication_error"}}, + ) + with pytest.raises(litellm.AuthenticationError, match="Invalid API key"): + litellm.completion( + model="iteracompute/qwen/qwen3.8-27b", + messages=[{"role": "user", "content": "Hello"}], + api_key="invalid-test-key", + ) + + +def test_responses_bridge_keeps_model_namespace() -> None: + with respx.mock() as upstream: + route: Final = upstream.post("https://api.iteracompute.com/v1/chat/completions").respond(200, json=_COMPLETION) + response: Final = litellm.responses(model="iteracompute/qwen/qwen3.8-27b", input="Hello", api_key="test-key") + + assert json.loads(route.calls.last.request.content)["model"] == "qwen/qwen3.8-27b" + assert response.output[0].content[0].text == "Hello" + + +@pytest.mark.asyncio +async def test_messages_bridge_keeps_model_namespace( + monkeypatch: pytest.MonkeyPatch, +) -> None: + monkeypatch.setattr(litellm, "disable_aiohttp_transport", True) + monkeypatch.setattr(litellm, "in_memory_llm_clients_cache", LLMClientCache()) + with respx.mock() as upstream: + route: Final = upstream.post("https://api.iteracompute.com/v1/chat/completions").respond(200, json=_COMPLETION) + response: Final = await litellm.anthropic.messages.acreate( + model="iteracompute/qwen/qwen3.8-27b", + messages=[{"role": "user", "content": "Hello"}], + max_tokens=512, + api_key="test-key", + ) + + assert json.loads(route.calls.last.request.content)["model"] == "qwen/qwen3.8-27b" + assert response["content"][0]["text"] == "Hello"