$0.20 / $0.66
Per 1M tokens (in / out)
Cached read $0.04 per 1M tokens. A fraction of the cost of frontier models.
LOW-COST REASONING MODEL
Omen Alphafor coding.For what comes next.
Widely reported as the 2.0 successor to Ox Alpha.
$0.20 in, $0.66 out per million tokens. Zero data retention. An OpenAI-compatible API for coding, agentic work and production use.
Official Pricing$0.20 in / $0.66 out
Tokenra Pricing$0.08 in / $0.26 out60% OFF
$0.20 / $0.66 per 1M tokens
0-day data retention
No training on your data
GET STARTED
Three steps to your first call.
Omen Alpha is served through the TokenRa route under the model ID omen-alpha. No setup fees — pay only for the tokens you use.
0d
Data retention. Your prompts are not used for training and are not stored.
1
Create an account at tokenra.io/register.
2
Generate an API key in your TokenRa dashboard.
3
Call the OpenAI-compatible endpoint with model: "omen-alpha".
GO DEEPER
Explore the model record.
Start with the answer you need, then follow the evidence, measurements, or setup details.
MODEL IDENTITY
What is Omen Alpha?
Separate confirmed facts from community evidence about its identity, lineage, and relationship to Ox Alpha.
Read the model recordMEASURED RESULTS
Omen Alpha benchmarks.
Review dated coding scores, observed cost and speed, methodology, and comparison context.
See the benchmark dataPRACTICAL GUIDE
How to use Omen Alpha.
Connect through Tokenra or OpenCode Go with the omen-alpha model identifier.
MODEL COMPARISON
Omen Alpha vs Ox Alpha.
Compare pricing, context, benchmark evidence, access, and practical use cases across both stealth models.
Read the comparisonMODEL SPECIFICATIONS
Frontier-scale pricing. Zero retention.
0 days
Data retention
Prompts are not used for training and are not retained.
omen-alpha
Model ID
OpenAI-compatible Chat Completions endpoint. Get a key at tokenra.io.
CODING BENCHMARK SNAPSHOT
A measured starting point.
23.14/40
OpenCode coding score
Recorded on September 4, 2026 across four coding projects.
#15
Leaderboard rank
Rank in the dated OpenCode leaderboard snapshot.
READ THE RESULT IN CONTEXT
Average cost was $0.03 per prompt and average time was 01:51. See the methodology before comparing models.
AT-A-GLANCE REFERENCE
Access details.
| Item | Details |
|---|---|
| Model identifier | omen-alpha in the chat completion request body. |
| API format | OpenAI-compatible Chat Completions API. |
| Endpoint | zen/go/v1/chat/completions (base URL: see TokenRa docs). |
| Pricing | $0.20 input / $0.66 output / $0.04 cached read, per 1M tokens. |
| Privacy | Not used for training. Data retention: 0 days. |
| Parameters | Not yet publicly disclosed. Community evaluation is ongoing. |
| Access | Create an account at tokenra.io/register, then generate an API key in your dashboard. |
FREQUENTLY ASKED
Straight answers.
The facts below reflect what is currently disclosed. Anything not listed — like parameter counts — has not been made public yet.
What is Omen Alpha?
Omen Alpha is a low-cost reasoning model widely reported as the 2.0 successor to Ox Alpha, available at omenalpha.io and served with the model ID omen-alpha. The relationship has not been officially confirmed.
How much does Omen Alpha cost?
$0.20 per 1M input tokens, $0.66 per 1M output tokens, and $0.04 per 1M cached-read tokens. Confirm live pricing on TokenRa before deploying.
Does Omen Alpha use my data for training?
No. Prompts are not used for training, and the stated data retention is 0 days.
How do I get API access?
Create an account at tokenra.io/register, generate an API key in your dashboard, and call the OpenAI-compatible endpoint with model: "omen-alpha". Keep production keys server-side — never in client-side code.
What is the context window?
Model parameters, including the context window, are not yet publicly disclosed. Check the official site and TokenRa listing for updates.
Is Omen Alpha the 2.0 version of Ox Alpha?
Multiple community sources report Omen Alpha as the 2.0 successor to Ox Alpha, but the relationship has not been officially confirmed by either team.