Flagship
Long-horizon coding
Sustained software engineering: debugging, refactoring, and reasoning across large codebases without losing context.
Reasoning model · free to try
A frontier reasoning model built for code, long-horizon agents, and a million tokens of context. Fast, capable, and free to try.
Research team · September 2026
Specification
Context window
tokens held in memory at once
Max output
tokens in a single response
Input modalities
text · image · video
Price to try
no account, no card
Reasoning
Ox Alpha is a reasoning-first model. It plans, checks itself, and only then answers — so hard problems get worked through instead of guessed at.
Tokens held in memory
Paste an entire codebase, a 500-page spec, or a week of transcripts. Ox Alpha keeps all of it in mind while it works — no chunking, no retrieval tricks, no losing the thread.
Capabilities
Everything a frontier model should do — and a few things most of them can't.
Flagship
Sustained software engineering: debugging, refactoring, and reasoning across large codebases without losing context.
Vision
Text, images, and video in. Screenshots, diagrams, and UI — reasoned about, not just described.
Agents
Native tools and tool_choice support for agents that actually do things.
Integration
Ask for JSON and get JSON — schema-shaped responses via response_format.
Stamina
Keeps its footing across long, multi-step tasks instead of drifting off halfway.
Access
Open the browser and start. No account, no card, no waitlist — on any device.
In action
A reasoning model shows its work. It names the failure mode before it writes the fix.
Prompt
Why does this Python function sometimes return stale data?
cache = {}
def get(k): return cache.get(k) or fetch(k)
Thinking
or treats falsy cached values (0, "", []) as misses and re-fetches. And nothing ever writes to cache — so the stale data must come from fetch itself.
Answer
def get(k):
if k not in cache: # explicit miss check
cache[k] = fetch(k) # actually populate
return cache[k]
Independent coding benchmark
8 of 10 real-world coding tasks solved — ahead of every frontier model it was lined up against.
Mean score, same 10 tasks
Reference models scored as passes out of 4 attempts per task; Ox Alpha as pass/fail.
Ten tasks, side by side
| Task | Fable 5 | GLM-5.3 | GPT-5.6 | Grok 4.6 | Ox Alpha |
|---|---|---|---|---|---|
| anko-typed-variable-bindings | 4/4 | 4/4 | 2/4 | 1/4 | ✓ |
| arktype-json-schema-refs | 2/4 | 1/4 | 3/4 | 1/4 | ✓ |
| fastapi-deprecation-headers | 4/4 | 3/4 | 3/4 | 4/4 | ✓ |
| helm-unified-manifest-stream | 4/4 | 4/4 | 4/4 | 4/4 | ✓ |
| igel-persist-feature-schema | 3/4 | 3/4 | 0/4 | 4/4 | ✓ |
| katex-multicolumn-array-spans | 2/4 | 4/4 | 3/4 | 4/4 | ✓ |
| meriyah-explicit-resource-decl | 1/4 | 0/4 | 0/4 | 0/4 | ✓ |
| query-persist-restored-state | 2/4 | 3/4 | 1/4 | 2/4 | ✓ |
| scc-bounded-memory-spilling | 4/4 | 3/4 | 4/4 | 4/4 | ✕ |
| vulture-persistent-analysis-cache | 0/4 | 0/4 | 1/4 | 1/4 | ✕ |
| Mean on these 10 | 65% | 62% | 52% | 62% | 80% |
Highlighted row: the one task every reference model scored 1/4 or worse on — Ox Alpha solved it. Third-party data, small sample — directional, not definitive.
How it works
Step 01
Type a goal in plain English. Paste the whole file, log, or document — context is not a problem.
Step 02
It thinks through the problem step by step, then streams back its answer in real time.
Step 03
Refine with follow-ups. The whole conversation stays in its 1M-token memory.
Try it on
Coding
Root-cause analysis from a model built for long-horizon engineering.
Reasoning
It plans before it answers — steps, not guesses.
Big context
1M tokens means the repo fits. No chunking, no RAG.
Vision
Text plus visual context, reasoned about together.
Agentic
Tool calling, structured output, stamina for long tasks.
Coding
Correctness, edge cases, simplifications — with reasoning.
Writing
Structured, precise writing tuned for production.
Big context
Summary, quotes, action items — whole document in context.
Access & privacy
Messages are processed by the upstream model API to generate responses. As with any AI service, avoid sharing sensitive data.
Free to try
No account, no card, no waitlist.
oxalpha.com · 2026