Arena

Pre-computed · no key needed

Compare 2 models side-by-side on a fixed prompt. Answers replay from real benchmark runs with TTFT and tokens/sec — no sign-up or API key required. Results updated Jul 30, 2026.

2/2 selected

Prompt

Prompt & models

Say hello in one short sentence.

Models

18 available

Models · 18

GPT-5.6 Sol (OpenAI)

openai/gpt-5.6-sol

Claude Opus 5 (Anthropic)

anthropic/claude-opus-5

Gemini 3.6 Flash (Google)

google/gemini-3.6-flash

Grok 4.5 (xAI)

x-ai/grok-4.5

DeepSeek V4 Pro (DeepSeek)

deepseek/deepseek-v4-pro

Qwen3.7 Flash (Qwen)

qwen/qwen3.7-flash

Kimi K3 (MoonshotAI)

moonshotai/kimi-k3

GLM 5.2 (Z.ai)

z-ai/glm-5.2

Mistral Medium 3.5 (Mistral)

mistralai/mistral-medium-3-5

Nemotron 3 Ultra (NVIDIA)

nvidia/nemotron-3-ultra-550b-a55b

GPT-5.6 Terra (OpenAI)

openai/gpt-5.6-terra

Claude Sonnet 5 (Anthropic)

anthropic/claude-sonnet-5

Gemini 3.5 Flash Lite (Google)

google/gemini-3.5-flash-lite

Grok 4.3 (xAI)

x-ai/grok-4.3

DeepSeek V4 Flash (DeepSeek)

deepseek/deepseek-v4-flash

Qwen3.7 Plus (Qwen)

qwen/qwen3.7-plus

Kimi K2.7 Code (MoonshotAI)

moonshotai/kimi-k2.7-code

GLM 5.1 (Z.ai)

z-ai/glm-5.1

Answers, latency, and throughput come from real runs captured ahead of time, then replayed here. Pick 2 models and hit Run.

GPT-5.6 SolReady

OpenAIopenai/gpt-5.6-sol

inter-token latencyno data

waiting for stream…

cumulative tokenstotal 0

waiting for stream…

tokens / secno data

waiting for stream…

—

Claude Opus 5Ready

Anthropicanthropic/claude-opus-5

inter-token latencyno data

waiting for stream…

cumulative tokenstotal 0

waiting for stream…

tokens / secno data

waiting for stream…

—

Powered by Antethink