Skip to main content

Apex Review: Callstack's React Native Coding Model at $0.50 per Million Tokens

Pick Apex for React Native or Next.js code at a low price; for broad general knowledge, Callstack says frontier models still do better.

Open the Playground

Spicy Index

77.7/ 100

Preliminary

Freedom Score

Not yet tested

Price
$1.13 per 1M mixed tokens (3 input : 1 output)
Output speed
141 tok/s
Success rate
Collecting data

Apex is a coding model from Callstack, the React Native consultancy, generally available since 1 October 2026. Callstack positions it for everyday agent work, with its focus on React Native and then Next.js and React for the web. On our API it is one OpenAI-compatible chat endpoint with a 262,144-token window, replies up to 131,072 tokens, tool calling and streaming, at $0.50 per million input tokens and $3 output.

The verdict

This is a catalogue review: we have not run our own quality or content tests. Callstack publishes 87.8% on its React Native evals against 88.6% for Claude Opus 5.5, and 96.8% on Next.js evals with docs in context (90.3% without). Those are Callstack's numbers, and Callstack says frontier models still do better on broad general-knowledge tasks.

That makes Apex a specialist: the interesting claim is near-frontier React Native results at a fraction of frontier prices, and it is one you should check on your own repository.

Try Apex →

Key facts

  • Tested by us: catalogue check and launch smoke run only, no quality or content tests; eval numbers are Callstack's
  • Price: $0.50 per million input tokens, $0.20 cached and $3 output
  • Limits: 262,144-token context, replies up to 131,072 tokens (default 4,096), up to 256 messages and 128 tools per request
  • Reasoning: none, low, medium or xhigh; medium is the default
  • Input: text; image parts are accepted but not documented by the publisher
  • Sources: the scoring rules are on our review methodology page, and the full parameter list is on the Apex model page.

What we liked and what held it back

What we liked

  • Specialised on React Native. Callstack trained it on React Native documentation, source and repository-grounded discussions.
  • Next.js and React too. Callstack publishes Next.js evals of 96.8% with docs in context.
  • A low price for a coding model: $0.50 in and $3 out.
  • Reasoning from none to xhigh so quick edits stay cheap.
  • Tools and streaming. Up to 128 function definitions, parallel calls on by default.

What held it back

  • A specialist. Callstack says frontier models are better on broad general-knowledge tasks.
  • Image input is unverified. The schema accepts image parts but Callstack lists text only.
  • No high or max level. The reasoning options are none, low, medium and xhigh.
  • No response_format field on this endpoint.
  • Evals are the publisher's. We have not reproduced them.

What we tested

We have not run Apex through our capability sets or content levels. This page uses Callstack's product pages and launch post (dated 1 October 2026), the limits and parameters of our endpoint and the live price on the model page. The Next.js evals are published on nextjs.org (31 tasks, pass@4, 25 September 2026); we have not rerun them.

Prompts that worked

Starting points based on the endpoint's shape, not prompts we benchmarked.

1. Review a list screen. The workload Callstack built the model for.

text
Review a React Native list screen for unnecessary renders and propose a focused patch.

Run this prompt on Apex →

2. A tool step. One function, tool_choice set to auto.

text
Call report_status with status ready.

Run this prompt on Apex →

Settings that matter: reasoning_effort none or low for quick edits and explanations, xhigh for a bug that spans several files; reasoning tokens count toward max_tokens, so raise it above 4,096; paste the relevant docs into context for Next.js work, since Callstack's higher Next.js number was with docs.

Roleplay and how far it writes

We have not tested Apex on our content levels yet, so this review makes no claim about how far it writes or how often it softens or refuses. The model page carries no content-test results for it either. On our API we add no content filter; whatever comes back is the model's own behaviour.

Not tested yet: all five content levels, roleplay quality over many turns, and languages other than English. We will add results here once the runs are done.

Every character in a request should be an adult, keep real, identifiable people out of it, and you're responsible for the law where you publish.

Alternatives

  • A broad frontier model? Claude Sonnet 5.5 is a measured option for general work.
  • Open weights, million-token window? Beam 1.0 is text-only at $1 in and $3 out.
  • Cheaper, long window? Claude Haiku 5.5 has a full review.

How Apex earns its Spicy Index

Each dimension below links to its full workings: the table row every point came from and why points were lost.

Public leaderboards 40% of quality
Not yet tested

No exact match on a tracked public leaderboard yet. How it works

Role-play 15% of quality
Not yet tested

Our role-play suite hasn’t run on this model yet. How it works

Creative writing 15% of quality
65.6
Instruction following 15% of quality
100.0
Long-context memory 15% of quality
Not yet tested

Our long-context suite hasn’t run on this model yet. How it works

Capability 10.5% weight
49.0
Freedom Score 30% weight
Not yet tested

Freedom evidence is untested or insufficient. The Index retains the original quality and capability score, is marked preliminary and still receives a rank. Missing evidence is not zero. How it works

Freedom evidence is untested or insufficient. The Index retains the original quality and capability score, is marked preliminary and still receives a rank. Missing evidence is not zero.

Open the full test report →

Capability, item by item

Points from the public scoring table. Upscaled tiers and limits we set ourselves don’t count.

Context window (creator-published)Not declared
0.0 / 30
Input modalities on SpicyAPIText, plus images
12.0 / 25
Tool useTools and tool_choice
15.0 / 15
Structured outputNot accepted
7.0 / 15
Reasoning control4 levels: none · low · medium · xhigh
15.0 / 15

What Apex can do on SpicyAPI

Straight from the live model specification on SpicyAPI. Some creators offer more in their own apps; this is what you can call through our API today.

Capabilities by endpoint
FeatureChat
Context window262,144 tokens
Max output per request131,072 tokens
Reasoning levelsnone · low · medium · xhigh
Tool callingYes
Structured outputNo
See every parameter and code sample →

How long it really takes

Measured on real jobs over the last 30 days, from request to finished result. Times include our own test runs.

Output speed
141 tok/sIncludes time to first token
Success rate
Collecting data

A time appears once at least 10 completed jobs are in. Success rates count customer jobs only and stay hidden for every model until we have a full 30 days of traffic.

Models to consider instead of Apex

What do you think?

Was this review useful?

Log in to vote.

Comments

Comments are checked automatically and by our team. We remove spam, offensive content, personal attacks and anything that names a real person without consent. Report a comment and we review it within one business day.

Tried this model? Share what worked. You need a free account to comment, so we can keep spam out.

Apex FAQ

Try Apex yourself

Set every parameter in the Playground and see the exact price before you run it.

$1.13 per 1M mixed tokens (3 input : 1 output)

More reviews

See the leaderboards →
  • Small blank marble tiles race along a short brass rail on a basalt plinth into a three-slot marble gate, where a gold filament splits into three channels ending in three neat stacks

    GPT 6.1 Terra

    Pick GPT 6.1 Terra for high-volume screens, documents and writing where reasoning_effort none helps; so far we have run only launch checks, no scored tests.

    Spicy Index
    79.9
    Freedom Score
    —

    Tested Oct 28, 2026

    Read the review
  • Small blank marble tiles race along a short brass rail on a basalt plinth into a three-slot marble gate, where a gold filament splits into three channels ending in three neat stacks

    Kimi K3.1 Pro

    Pick Kimi K3.1 Pro for long-context work with screenshots and clips when you accept always-on thinking; so far we have run only launch checks, no scored tests.

    Spicy Index
    80.4
    Freedom Score
    —

    Tested Oct 28, 2026

    Read the review
  • A burst of golden light filaments fanning low across a shallow glass dish against pure black

    Grok 4.3

    Pick Grok 4.3 for a huge context window, a genuine reasoning off switch and uncensored writing at a low price; skip it if you need file uploads.

    Spicy Index
    77.0
    Freedom Score
    98.9

    Tested Sep 24, 2026

    Read the review