Venice Models API

Venice Models

Three read-only endpoints for model discovery - all GET, all public:

Auth is optional. These routes need no API key, so a plain curl works. If you send Authorization: Bearer $VENICE_API_KEY the result is tailored to that caller: a key's modelPrivacy setting filters the list, beta-flagged accounts also see beta models, and partner accounts see their negotiated rates. An invalid key is not rejected - you just get the public view.

?type= values (default text when omitted):

Anything else returns 400 { "error": "Invalid request parameters", "details": …, "issues": … }.

music includes long-form audio (songs, sound effects, ElevenLabs narration models) and any voice-changer models. decision lists the typed-judgment models used by POST /decisions (today: jev-latest, flagged betaModel: true) - see venice-decisions.

Use when

• You need to pick a model at runtime based on capabilities (vision, reasoning, function calling, E2EE, X search, multi-image, …).
• You need to validate a request against a model's constraints (prompt length, aspect ratio, resolution, steps, durations).
• You need the current price per million tokens / per image / per second / per million characters to build a cost estimate.
• You want to resolve a trait (default, default_reasoning, highest_quality) or a legacy alias (gpt-4o) to a concrete Venice model ID.

GET /models

curl "https://api.venice.ai/api/v1/models?type=text"

{
  "object": "list",
  "type": "text",
  "data": [
    {
      "id": "zai-org-glm-5-2",
      "object": "model",
      "owned_by": "venice.ai",
      "type": "text",
      "created": 1781568000,
      "context_length": 1000000,
      "model_spec": {
        "name": "GLM 5.2",
        "description": "GLM-5.2 is the next-generation large language model…",
        "availableContextTokens": 1000000,
        "maxCompletionTokens": 131072,
        "privacy": "private",
        "modelSource": "https://huggingface.co/zai-org/GLM-5.2",
        "offline": false,
        "traits": ["default", "function_calling_default"],
        "capabilities": { "supportsReasoning": true, "reasoningEffortOptions": ["none", "high", "max"], "…": "…" },
        "pricing": {
          "input":       { "usd": 1.4,  "diem": 1.4 },
          "cache_input": { "usd": 0.26, "diem": 0.26 },
          "output":      { "usd": 4.4,  "diem": 4.4 }
        }
      }
    }
  ]
}

Top-level ModelResponse fields

model_spec - common fields

model_spec.capabilities - text models

model_spec.constraints and type-specific fields

• Text - constraints is optional (only a handful of models carry it): temperature.default, top_p.default, optional {frequency,presence,repetition}_penalty.default.
• Image - constraints: promptCharacterLimit, widthHeightDivisor, steps.{default,max}, optional aspectRatios[] + defaultAspectRatio, optional resolutions[] + defaultResolution, optional qualities[] + defaultQuality (models that accept quality), optional maxStyleReferences + supportsStyleReferenceStrength. Alongside: supportsStyleReferences, plus supportsWebSearch and supportsOptimizePromptThinking (served live, not in the OpenAPI schema).
• Inpaint / edit - constraints: aspectRatios[], promptCharacterLimit, combineImages, optional maxInputImages, singleImageAspectRatio (if false, single-image edits keep input dimensions and ignore aspect_ratio), optional resolutions[]/defaultResolution, qualities[]/defaultQuality. Alongside: supportsOptimizePromptThinking.
• Video - constraints: model_type (text-to-video / image-to-video / video), aspect_ratios[], resolutions[], durations[] (e.g. "5s"; most upscale and video-to-video models list "Auto", meaning the source length — omit duration for them), audio, audio_configurable, audio_input, per_reference_audio, video_input, optional prompt_character_limit (default 2500), optional reference_image_min_short_side_pixels, reference_image_min_aspect_ratio, reference_image_max_aspect_ratio, and a topaz block (models, sliders, selects, no_upscale_models, h264_output, prompt) on enhancement models only. The audio_input … reference_image_* keys are served live but not declared in the OpenAPI schema.
• TTS (top level of model_spec) - voices[], default_format, supported_formats[] (an explicit format outside this list is rejected), supports_custom_voice_id, and voice_cloning { mode: "zero_shot" | "persistent", accepted_formats[], min_sample_seconds, retention_days } on models whose cloning is open to you (use with POST /audio/voices, see venice-audio-speech). Per-model toggles like prompt / temperature / top_p support are not exposed here - use the per-model table in venice-audio-speech as the support matrix (the published schema text is incomplete).
• Music / audio generation (top level) - supports_lyrics, lyrics_required, supports_force_instrumental, supports_lyrics_optimizer (served live, not in the OpenAPI schema), supports_loop, supports_custom_voice_id, supports_language_code, supports_speed, supported_formats[], default_format, prompt_character_limit, min_prompt_length, optional lyrics_character_limit, duration_options[], min_duration / max_duration / default_duration, voices[] / default_voice, default_speed / min_speed / max_speed.
• Voice changer (music models with voice_changer: true) - adds supports_background_noise_removal, supports_seed, accepted_audio_formats[], max_source_audio_duration_seconds. These run on /audio/voice-changer/*, not /audio/queue - see venice-audio-voice-changer. No voice-changer model is publicly listed today; check ?type=music for voice_changer: true before relying on it.
• Embedding (top level) - embeddingDimensions, maxInputTokens, supportsCustomDimensions (present only when true).
• Decision (top level) - maxStateTokens (state + longest question), maxTotalTokens (state + all questions).
• ASR / upscale - no extra fields beyond the common ones and pricing.