Coding Credits · 3 hero models · limited launch

Claude Opus 4.7 + Sonnet 4.6 + GPT-5.5 — up to 7.7× value

Prepay credits, plug into Claude Code / Codex / Cursor, and every credit goes 7.7× further. No subscription, no monthly cap.

up to 7.7× valuePay as you go · no subscriptionCredits expire 365dCard · Crypto

Claude Opus 4.7Anthropic flagship

7.7×value

Claude Sonnet 4.6Workhorse

7.7×value

GPT-5.5OpenAI mainline

5×value

Pick a pack

Which credits get consumed?

One API key works for both. Routing is decided by the model you call.

Coding models

Claude Opus 4.7, Sonnet 4.6, GPT-5.5 drain Coding Credits first (FIFO), then spill to General Credits if Coding Credits run out.

Every other model

Image, video, audio, 3D, and all other LLMs drain General Credits only. Coding Credits are never touched — enforced at the SQL layer.

Pick a pack

One-time purchase. Credits expire 365 days after purchase.

coding-starter

Bonus +25%

$20

2,500 coding credits

2,000 + 500 bonus

Real value

5.5×

coding-pro

Bonus +50%

$200

30,000 coding credits

20,000 + 10,000 bonus

Real value

6.6×

coding-max

Bonus +75%

$1000

175,000 coding credits

100,000 + 75,000 bonus

Real value

7.7×

Credits expire 365 days after purchase. Coding credits are locked to the 3 eligible models — they cannot be spent on video / image / other LLMs.

Which models do these credits work on?

Exactly three model families. Hard-coded — not configurable.

Claude Opus 4.7 / 4.6

Claude Sonnet 4.6

GPT-5.5

Coding credits are sandboxed at the SQL layer. If a request hits any other model (video, image, audio, 3D, or non-Hero LLMs), the deduction function refuses to touch your coding balance. Your general credits handle that — coding credits stay protected for what you bought them for.

How far each credit goes

Same upstream models. Same tokens. A lot more value per dollar.

Model	Our $ / MTok (in / out)	Public API rate	Your multiplier
Claude Opus 4.7 / 4.6	$3.40 / $16.96	$15.00 / $75.00	−77%
Claude Sonnet 4.6	$0.68 / $3.40	$3.00 / $15.00	−77%
GPT-5.5	$0.42 / $2.52	$1.20 / $7.20	−65%

Per 1M tokens. Output dominates real coding workloads — most of the value shows up there.

How to use it

Drop Hypereal into any tool that speaks OpenAI or Anthropic-compatible APIs.

Claude Code

The official Anthropic Claude Code CLI talks to /v1/messages — point it at Hypereal and pin Opus or Sonnet.

# 1. Get your API key
#    https://hypereal.cloud/manage-api-keys

# 2. Point Claude Code at Hypereal (Anthropic-compatible endpoint)
export ANTHROPIC_BASE_URL="https://hypereal.cloud/v1"
export ANTHROPIC_API_KEY="sk-hyp-..."   # your Hypereal key

# 3. Pin a coding-plan model (opus 4.7 or sonnet 4.6)
export ANTHROPIC_MODEL="claude-opus-4-7"

# 4. Run
claude

OpenAI Codex CLI

OpenAI Codex CLI uses /v1/chat/completions. Set OPENAI_BASE_URL to Hypereal and pin GPT-5.5.

export OPENAI_BASE_URL="https://hypereal.cloud/v1"
export OPENAI_API_KEY="sk-hyp-..."
export OPENAI_MODEL="gpt-5.5"

codex --provider openai

Cursor

Cursor accepts any OpenAI-compatible endpoint. Override the OpenAI host in Settings → Models.

# Cursor → Settings → Models → "Add OpenAI-compatible"
#   Base URL:  https://hypereal.cloud/v1
#   API Key:   sk-hyp-...
#   Model:     gpt-5.5         (or claude-opus-4-7 / claude-sonnet-4-6)
#
# Enable "Override OpenAI" in the same panel so Cursor routes
# completions through Hypereal instead of api.openai.com.

OpenCode / Cline / Aider

OpenCode, Cline (VS Code), Aider — all of them accept ANTHROPIC_API_BASE / API_KEY env vars or settings.

# OpenCode / Cline (VS Code) — Anthropic-compatible
{
  "anthropic.baseURL": "https://hypereal.cloud/v1",
  "anthropic.apiKey":  "sk-hyp-...",
  "anthropic.model":   "claude-opus-4-7"
}

# Aider — pass through env
export ANTHROPIC_API_BASE="https://hypereal.cloud/v1"
export ANTHROPIC_API_KEY="sk-hyp-..."
aider --model anthropic/claude-opus-4-7

Hermes Agent

Hermes Agent works with any OpenAI-compatible endpoint. Configure once and run.

# Hermes Agent supports OpenAI-compatible endpoints out of the box.
hermes config set provider openai-compatible
hermes config set base_url https://hypereal.cloud/v1
hermes config set api_key sk-hyp-...
hermes config set model gpt-5.5

hermes run "Refactor auth.ts to use jose"

Raw curl

For agents and scripts: hit /v1/messages directly with your API key.

curl https://hypereal.cloud/v1/messages \
  -H "x-api-key: sk-hyp-..." \
  -H "content-type: application/json" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-opus-4-7",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Write a Python quicksort."}]
  }'

FAQ

Can I spend these credits on video or image models?

No. Coding credits are locked to Claude Opus 4.7 / 4.6, Claude Sonnet 4.6, and GPT-5.5. Any other model deducts from your general balance instead.

Do credits expire?

365 days after purchase. Set up auto top-up in Settings → Credits if you do not want to think about it.

Refund policy?

Unused credits can be refunded within 14 days. Used credits are non-refundable. See /refund-policy.

Are there rate limits?

No monthly quota and no artificial throttle. You pay per token like the upstream API would charge — just at our price.

Is this a subscription?

No. One-time prepaid packs. You decide when to top up. No auto-renew, no surprise charges.

Coding Credits · 3 hero models · limited launch

Claude Opus 4.7 + Sonnet 4.6 + GPT-5.5 — up to 7.7× value

Prepay credits, plug into Claude Code / Codex / Cursor, and every credit goes 7.7× further. No subscription, no monthly cap.

up to 7.7× valuePay as you go · no subscriptionCredits expire 365dCard · Crypto

Claude Opus 4.7Anthropic flagship

7.7×value

Claude Sonnet 4.6Workhorse

7.7×value

GPT-5.5OpenAI mainline

5×value

Pick a pack

Which credits get consumed?

One API key works for both. Routing is decided by the model you call.

Coding models

Claude Opus 4.7, Sonnet 4.6, GPT-5.5 drain Coding Credits first (FIFO), then spill to General Credits if Coding Credits run out.

Every other model

Image, video, audio, 3D, and all other LLMs drain General Credits only. Coding Credits are never touched — enforced at the SQL layer.

Pick a pack

One-time purchase. Credits expire 365 days after purchase.

coding-starter

Bonus +25%

$20

2,500 coding credits

2,000 + 500 bonus

Real value

5.5×

coding-pro

Bonus +50%

$200

30,000 coding credits

20,000 + 10,000 bonus

Real value

6.6×

coding-max

Bonus +75%

$1000

175,000 coding credits

100,000 + 75,000 bonus

Real value

7.7×

Credits expire 365 days after purchase. Coding credits are locked to the 3 eligible models — they cannot be spent on video / image / other LLMs.

Which models do these credits work on?

Exactly three model families. Hard-coded — not configurable.

Claude Opus 4.7 / 4.6

Claude Sonnet 4.6

GPT-5.5

How far each credit goes

Same upstream models. Same tokens. A lot more value per dollar.

Model	Our $ / MTok (in / out)	Public API rate	Your multiplier
Claude Opus 4.7 / 4.6	$3.40 / $16.96	$15.00 / $75.00	−77%
Claude Sonnet 4.6	$0.68 / $3.40	$3.00 / $15.00	−77%
GPT-5.5	$0.42 / $2.52	$1.20 / $7.20	−65%

Per 1M tokens. Output dominates real coding workloads — most of the value shows up there.

How to use it

Drop Hypereal into any tool that speaks OpenAI or Anthropic-compatible APIs.

Claude Code

The official Anthropic Claude Code CLI talks to /v1/messages — point it at Hypereal and pin Opus or Sonnet.

# 1. Get your API key
#    https://hypereal.cloud/manage-api-keys

# 2. Point Claude Code at Hypereal (Anthropic-compatible endpoint)
export ANTHROPIC_BASE_URL="https://hypereal.cloud/v1"
export ANTHROPIC_API_KEY="sk-hyp-..."   # your Hypereal key

# 3. Pin a coding-plan model (opus 4.7 or sonnet 4.6)
export ANTHROPIC_MODEL="claude-opus-4-7"

# 4. Run
claude

OpenAI Codex CLI

OpenAI Codex CLI uses /v1/chat/completions. Set OPENAI_BASE_URL to Hypereal and pin GPT-5.5.

export OPENAI_BASE_URL="https://hypereal.cloud/v1"
export OPENAI_API_KEY="sk-hyp-..."
export OPENAI_MODEL="gpt-5.5"

codex --provider openai

Cursor

Cursor accepts any OpenAI-compatible endpoint. Override the OpenAI host in Settings → Models.

# Cursor → Settings → Models → "Add OpenAI-compatible"
#   Base URL:  https://hypereal.cloud/v1
#   API Key:   sk-hyp-...
#   Model:     gpt-5.5         (or claude-opus-4-7 / claude-sonnet-4-6)
#
# Enable "Override OpenAI" in the same panel so Cursor routes
# completions through Hypereal instead of api.openai.com.

OpenCode / Cline / Aider

OpenCode, Cline (VS Code), Aider — all of them accept ANTHROPIC_API_BASE / API_KEY env vars or settings.

# OpenCode / Cline (VS Code) — Anthropic-compatible
{
  "anthropic.baseURL": "https://hypereal.cloud/v1",
  "anthropic.apiKey":  "sk-hyp-...",
  "anthropic.model":   "claude-opus-4-7"
}

# Aider — pass through env
export ANTHROPIC_API_BASE="https://hypereal.cloud/v1"
export ANTHROPIC_API_KEY="sk-hyp-..."
aider --model anthropic/claude-opus-4-7

Hermes Agent

Hermes Agent works with any OpenAI-compatible endpoint. Configure once and run.

# Hermes Agent supports OpenAI-compatible endpoints out of the box.
hermes config set provider openai-compatible
hermes config set base_url https://hypereal.cloud/v1
hermes config set api_key sk-hyp-...
hermes config set model gpt-5.5

hermes run "Refactor auth.ts to use jose"

Raw curl

For agents and scripts: hit /v1/messages directly with your API key.

curl https://hypereal.cloud/v1/messages \
  -H "x-api-key: sk-hyp-..." \
  -H "content-type: application/json" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-opus-4-7",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Write a Python quicksort."}]
  }'

FAQ

Can I spend these credits on video or image models?

No. Coding credits are locked to Claude Opus 4.7 / 4.6, Claude Sonnet 4.6, and GPT-5.5. Any other model deducts from your general balance instead.

Do credits expire?

365 days after purchase. Set up auto top-up in Settings → Credits if you do not want to think about it.

Refund policy?

Unused credits can be refunded within 14 days. Used credits are non-refundable. See /refund-policy.

Are there rate limits?

No monthly quota and no artificial throttle. You pay per token like the upstream API would charge — just at our price.

Is this a subscription?

No. One-time prepaid packs. You decide when to top up. No auto-renew, no surprise charges.