AI Models

All models available through CorvusLLM — at up to 65% below official API pricing. Click any card for detailed specs, use-cases, and setup guidance.

All Models

Supported model catalog

Every model below is available through CorvusLLM's OpenAI-compatible API proxy. You connect once; switch models by changing a single parameter.

Anthropic - Claude

Anthropic Claude Haiku 4.5

Fast Claude tier for high-throughput automation, summarisation, extraction, and lightweight assistant work.

Official input: $1.00/1M Official output: $5.00/1M Cache read: $0.100/1M
Public slug claude-haiku-4-5
Fast Verified pricing OpenAI-compatible
Anthropic Claude Opus 4.5

Highest-capability Claude tier for deep reasoning, multi-step coding, and long research workflows when quality matters more than speed.

Official input: $5.00/1M Official output: $25.00/1M Cache read: $0.500/1M
Public slug claude-opus-4-5
Max capability Verified pricing OpenAI-compatible
Anthropic Claude Opus 4.5 Thinking

Highest-capability Claude tier for deep reasoning, multi-step coding, and long research workflows when quality matters more than speed.

Official input: $5.00/1M Official output: $25.00/1M Cache read: $0.500/1M
Public slug claude-opus-4-5-thinking
Thinking Max capability Verified pricing OpenAI-compatible
Anthropic Claude Opus 4.6

Highest-capability Claude tier for deep reasoning, multi-step coding, and long research workflows when quality matters more than speed.

Official input: $5.00/1M Official output: $25.00/1M Cache read: $0.500/1M
Public slug claude-opus-4-6
Max capability Verified pricing OpenAI-compatible
Anthropic Claude Opus 4.6 Thinking

Highest-capability Claude tier for deep reasoning, multi-step coding, and long research workflows when quality matters more than speed.

Official input: $5.00/1M Official output: $25.00/1M Cache read: $0.500/1M
Public slug claude-opus-4-6-thinking
Thinking Max capability Verified pricing OpenAI-compatible
Anthropic Claude Opus 4.7

Highest-capability Claude tier for deep reasoning, multi-step coding, and long research workflows when quality matters more than speed.

Official input: $5.00/1M Official output: $25.00/1M Cache read: $0.500/1M
Public slug claude-opus-4-7
Max capability Verified pricing OpenAI-compatible
Anthropic Claude Sonnet 4.5

Balanced Claude tier for day-to-day coding, agent loops, writing, and analysis with a strong quality-to-speed tradeoff.

Official input: $3.00/1M Official output: $15.00/1M Cache read: $0.300/1M
Public slug claude-sonnet-4-5
Balanced Verified pricing OpenAI-compatible
Anthropic Claude Sonnet 4.5 Thinking

Balanced Claude tier for day-to-day coding, agent loops, writing, and analysis with a strong quality-to-speed tradeoff.

Official input: $3.00/1M Official output: $15.00/1M Cache read: $0.300/1M
Public slug claude-sonnet-4-5-thinking
Thinking Balanced Verified pricing OpenAI-compatible
Anthropic Claude Sonnet 4.6

Balanced Claude tier for day-to-day coding, agent loops, writing, and analysis with a strong quality-to-speed tradeoff.

Official input: $3.00/1M Official output: $15.00/1M Cache read: $0.300/1M
Public slug claude-sonnet-4-6
Balanced Verified pricing OpenAI-compatible
Anthropic Claude Sonnet 4.6 Thinking

Balanced Claude tier for day-to-day coding, agent loops, writing, and analysis with a strong quality-to-speed tradeoff.

Official input: $3.00/1M Official output: $15.00/1M Cache read: $0.300/1M
Public slug claude-sonnet-4-6-thinking
Thinking Balanced Verified pricing OpenAI-compatible

OpenAI - GPT

OpenAI GPT 5.2

Balanced GPT tier for daily coding and chat workloads when you want dependable output at a lower blended cost.

Official input: $1.75/1M Official output: $14.00/1M Cache read: $0.175/1M
Public slug gpt-5.2
Verified pricing OpenAI-compatible
OpenAI GPT 5.3

GPT 5.3 is available through CorvusLLM's shared catalog with the same public slug everywhere the product surfaces it.

Official input: $1.75/1M Official output: $14.00/1M Cache read: $0.175/1M
Public slug gpt-5.3
Verified pricing OpenAI-compatible
OpenAI GPT 5.4

Flagship GPT tier exposed through CorvusLLM for strong coding, analysis, and general assistant workloads with a familiar OpenAI-compatible shape.

Official input: $2.50/1M Official output: $15.00/1M Cache read: $0.250/1M
Public slug gpt-5.4
Verified pricing OpenAI-compatible
OpenAI GPT 5.5

GPT 5.5 is available through CorvusLLM's shared catalog with the same public slug everywhere the product surfaces it.

Official input: $5.00/1M Official output: $30.00/1M Cache read: $0.500/1M
Public slug gpt-5.5
Verified pricing OpenAI-compatible

Z.AI - GLM

Z.AI GLM 5.1

Latest GLM tier with a sharper quality-to-price balance for general reasoning, coding, and multilingual assistant workloads.

Official input: $1.40/1M Official output: $4.40/1M Cache read: $0.260/1M
Public slug glm-5.1
Cost efficient Verified pricing OpenAI-compatible
Z.AI GLM 5

Cost-efficient GLM tier for general chat, automation, and production traffic where low blended pricing is important.

Official input: $1.00/1M Official output: $3.20/1M Cache read: $0.200/1M
Public slug glm-5
Cost efficient Verified pricing OpenAI-compatible

Interactive directory