Overview

CallMissed provides access to a tiered model catalog through a single OpenAI-compatible API:

• Fast LLMs — Kimi K2.5 at 414 tokens/second on Nvidia B200 GPUs. The default and fastest LLM for voice agents.
• Indic Models — purpose-built for Indian languages. STT, TTS, and LLM optimized for Hindi, Tamil, Telugu, Bengali, and 19 more.
• Direct-Routed LLMs — sub-2s open-weights models (Kimi K2.5/K2.6, GPT-OSS, Gemma-4, GLM, Nemotron, Mistral Small).
• Frontier Models — 300+ models from every major creator (OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, and more) through one endpoint.

All models use the same authentication and request format. Just change the model field.

Models API

List all available models programmatically. No authentication required.

# List all models
curl https://api.callmissed.com/api/v1/models

# Filter by category: llm, stt, tts
curl https://api.callmissed.com/api/v1/models?category=llm

# Filter free-plan models only
curl https://api.callmissed.com/api/v1/models?free=true

# Get a specific model
curl https://api.callmissed.com/api/v1/models/sarvam-30b

# Which models each plan tier can call
curl https://api.callmissed.com/api/v1/models/access

Response includes: id, name, description, category, owned_by, context_window, context_length (alias of context_window for OpenAI-style clients), pricing, free, supports_streaming, supports_tools, supports_reasoning, and supports_vision.

The OpenAI-compatible listing at GET /v1/models (requires Authorization: Bearer cm_*) returns the same enriched fields, and the Anthropic-shape listing at GET /anthropic/v1/models surfaces them inside Anthropic's {data, has_more, first_id, last_id} envelope.

Free Plan Models

The free tier includes 17 models across four categories. Use GET /api/v1/models?free=true to list them, or see the Model Access by Plan page for the full breakdown.

LLM (8 models)

STT (3 models)

TTS (4 models)

Image (6 models)

All other models (e.g. kimi-k2.5-fast, openai/*, anthropic/*, google/*, x-ai/*, qwen/*, mistralai/*) require a paid plan (Starter, Pro, or Enterprise).

Pricing

All models are pay-per-use. Pricing is in USD.

Full pricing for all models is available via the API: GET /api/v1/models

import requests

# List all LLM models
models = requests.get("https://api.callmissed.com/api/v1/models?category=llm").json()
for m in models["data"]:
    print(f"{m['id']} — {m['name']} ({m['context_window']} tokens) {'FREE' if m['free'] else 'PAID'}")

Fast LLMs

High-throughput Kimi K2.5 inference tier optimized for voice-agent latency.

While kimi-k2.5-fast is in maintenance (returns HTTP 503), use kimi-k2.5:

response = client.chat.completions.create(
    model="kimi-k2.5",
    messages=[{"role": "user", "content": "Hello"}]
)

Indic Models

Speech to Text

For 99-language general-purpose transcription, see whisper-large-v3-turbo. For diarization + smart-format on calls, see nova-3. Both are free-tier and live under the audio model routes.

Text to Speech

For low-latency English / Spanish voice agents, see aura-2-en / aura-2-es. For ultra-cheap en/fr notification audio, see melotts. All three are free-tier.

Chat Completion (LLM)

Both sarvam-30b and sarvam-105b support hybrid thinking mode via reasoning_effort: "low" | "medium" | "high". "none" and "minimal" are mapped down to "low" (verified 2026-05-01) so OpenAI-style clients sending reasoning_effort: "none" for thinking-off still get a 200. Full thinking-disable is available on the direct-routed kimi-k2.5 / kimi-k2.6 / gemma-4-26b-a4b-it models.

Audio Models

Free-tier on every plan. Source-of-truth pricing lives in backend/app/services/credit_service.py.

Speech to Text

Text to Speech

[Inference] Aura 2 returns linear16 PCM streamed at 24 kHz; LiveKit's voice agent plays it without an MP3 decode in the hot path. MeloTTS returns base64 MP3 and is decoded by pyav. AI behavior is not guaranteed and may vary as upstreams update their schemas.

Direct-Routed LLMs

Low-latency models routed directly through CallMissed — sub-2s end-to-end on small prompts and free-tier eligible per the reasoning_effort matrix.

Frontier Models

Access 300+ frontier models via the same /v1/chat/completions endpoint. Use the slash-prefixed model ID as the model field.

Popular Models

Use auto to let CallMissed select the best free model for your prompt automatically.

Model Selection

Pass the model ID in your request:

# Sarvam LLM
response = client.chat.completions.create(
    model="sarvam-30b",
    messages=[{"role": "user", "content": "Hello in Hindi"}]
)

# Indic LLM with thinking mode
response = client.chat.completions.create(
    model="sarvam-105b",
    messages=[{"role": "user", "content": "Solve this step by step"}],
    extra_body={"reasoning_effort": "high"}
)

# Frontier model
response = client.chat.completions.create(
    model="openai/gpt-5.4-mini",
    messages=[{"role": "user", "content": "Hello"}]
)

The API automatically routes to the correct backend based on the model ID:

• Bare names (kimi-k2.5, kimi-k2.6, gpt-oss-120b, gemma-4-26b-a4b-it, mistral-small-3.1, …) → direct-routed
• sarvam-* prefix → Indic LLMs
• Slash-prefixed (openai/, anthropic/, google/, …) → frontier catalog