Ship features
Plan and write changes across a codebase, then fix the tests.
Ox Alpha is a reasoning model built for coding, sustained agentic work and production workloads. Long-horizon software engineering, complex reasoning, and text-with-vision — right in your browser.
No signup · No VPN · No card

One model, many jobs
Ox Alpha combines deep reasoning with a huge context window and native tool use — the same model behind serious agentic and coding workloads.
Step-by-step thinking for math, logic and multi-constraint problems that shallow models miss.
01Reads whole codebases, plans changes across files and iterates on failures until it works.
02Built for sustained, long-horizon tasks with tool calling and multi-step execution.
03Feed entire repositories, books or document sets and keep the whole picture in view.
04Combines text with visual context — screenshots, diagrams and images.
05Reliable JSON and schema-constrained responses for production pipelines.
06What people build with it
Plan and write changes across a codebase, then fix the tests.
Reason over long documents and synthesise clear answers.
Drive multi-step tool workflows that run to completion.
Turn messy inputs into structured, actionable output.
Long-form writing with a consistent, context-aware voice.
Explain screenshots, diagrams and UI mockups in context.
Numbers, not adjectives
Routed directly through OpenRouter to a single dedicated provider — no routing guesswork, healthy failover.
How it stacks up
How Ox Alpha lines up against today's leading models on the things that matter — with one number no one else can match.
| Model | Context | Intelligence | Vision | Tools & agents | Price / 1M | Free |
|---|---|---|---|---|---|---|
Ox AlphaThis one | 1M | 59* | $0 | |||
Claude Opus 5 | 1M | 63 | $25 | |||
Claude Fable 5 | 1M | 62 | $50 | |||
Grok 4.6 | 500K | 61 | $6 | |||
GPT-5.6 Sol | 400K | 59 | $15 | |||
Gemini 3.7 Flash | 1M | 50 | $3.75 |
Indicative figures from public benchmarks (Artificial Analysis Intelligence Index, SWE-bench) as of August 2026. Ox Alpha is in preview — its score (*) is a community-reported estimate. Competitor prices are per 1M output tokens; Intelligence is the Artificial Analysis Intelligence Index.
Trusted by serious workloads
Ox Alpha powers demanding, high-volume applications — a good signal for what it can do for you.
Self-improving agents with memory, tools and browser automation.
Agentic tools that read your codebase and execute changes.
Long-context synthesis over large document sets.
Structured, schema-constrained output at scale.
The simplest plan there is
During the preview, every request to Ox Alpha is free — no card, no quotas to worry about.
Everything, right now
For teams — coming soon
No install, no extension. Open the chat on any device and start.
Early reactions
“Handed it a gnarly refactor across a dozen files and it just… did it. Free is wild.”
“The 1M context is the real deal — I dropped a whole spec in and it kept everything straight.”
“Fast enough for agent loops, cheap enough to leave running. My default now.”
“Reasoning quality feels frontier. No login, no VPN, opened it and went.”
Ox Alpha is a frontier reasoning model designed for coding, sustained agentic work and production workloads, with a 1M token context window and multimodal (text + vision) input.
Yes. During the preview every request is free — no card and no signup required to start chatting.
No. Ox Alpha runs in your browser. Just open the chat and type.
Up to 1,000,000 tokens in a single request — enough for entire codebases or large document sets.
Yes. Ox Alpha supports tool calling and schema-constrained structured outputs, which is why it powers agentic and production workloads.
Ox Alpha is multilingual and this site is available in 18 languages. Chat in whatever language you prefer.
Open the chat and put a frontier model to work — free, no strings.