CallMissed
Back to all models
LLM Chatfast

Kimi K2.5 Fast

by Moonshot · Released January 2026

The fast inference variant of Kimi K2.5. Same 1T parameter architecture but optimized for low-latency responses. Ideal for real-time applications where speed is critical.

LLM Chat

Kimi K2.5 Fast

Powered by Moonshot · Sparse Mixture-of-Experts (1T total / 32B active, optimized)

Context Window

128K

Parameters

1T total / 32B active (MoE)

Max Output

16K

Category

LLM Chat

Overview

Kimi K2.5 Fast is the speed-optimized variant of Moonshot AI's flagship K2.5 model. It shares the same 1 trillion total parameter MoE architecture with 32B active parameters per token, but is specifically optimized for low-latency inference — achieving 414 tokens per second throughput via Clarifai, making it one of the fastest large-scale models available.

The optimization focuses on inference efficiency without significantly compromising output quality. The model retains the core capabilities of K2.5 — native multimodality, strong coding performance, and flexible operating modes — while delivering responses fast enough for real-time chat applications, voice agent backends, and interactive coding assistants.

Kimi K2.5 Fast is the recommended choice when you need K2.5-level capability but latency is a primary concern. It handles real-time conversational AI, interactive coding sessions, and voice-driven applications where users expect near-instant responses.

Pricing

MetricPrice
Input /1M tokens₹52.0000
Output /1M tokens₹230.0000

1 credit = ₹1 = $0.01 USD. Prices shown from provider; CallMissed passes through with ~35% markup.

Key Highlights

  • Optimized for low-latency inference
  • Same architecture as K2.5 standard
  • 414 tokens/second throughput via Clarifai
  • Ideal for real-time chat applications

Benchmarks

BenchmarkScore
SWE-bench75.2%
LiveCodeBench83.8%
HumanEval90.7%
Throughput414 tok/s

Technical Details

  • Same 1T total / 32B active MoE architecture as K2.5 standard
  • Optimized for low-latency inference: 414 tokens/second via Clarifai
  • Inference optimizations include quantization and speculative decoding
  • Retains native multimodal capabilities from K2.5
  • Context window: 128K tokens
  • Open-weight model — same weights as K2.5 with optimized serving
  • Available via Moonshot API and CallMissed unified gateway

Strengths

  • 414 tok/s throughput — among the fastest large-scale models
  • Same architecture as K2.5 with minimal quality trade-off
  • Ideal for real-time and voice-driven applications
  • Open-weight — can be self-hosted with optimized serving

Limitations

  • Slight quality reduction compared to K2.5 standard on complex tasks
  • Same pricing as K2.5 standard — speed optimization, not cost optimization
  • 128K context is smaller than 1M-context competitors

Use Cases

Real-time chatVoice agent backendsLow-latency applicationsInteractive coding

API Example

curl https://api.callmissed.com/v1/chat/completions \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -d '{"model": "kimi-k2.5-fast", "messages": [{"role": "user", "content": "Quick answer: what is the capital of France?"}]}'

Endpoint: POST /v1/chat/completions · Model ID: kimi-k2.5-fast

Try Kimi K2.5 Fast now

Get 1000 free API credits on signup. No credit card required.

Start free Read docs