Pricing comparison

OpenAI vs Anthropic Pricing

OpenAI is cheaper than Anthropic at the entry level: gpt-oss-20b costs $0.140 per 1M output tokens against $1.25 for Claude 3 Haiku — 8.9× the price. On current flagships, OpenAI's GPT-5.6 costs $30.00/1M output against $50.00/1M for Anthropic's Claude Fable 5. Prices are USD, current as of August 11, 2026.

Per-million-token pricing for OpenAI and Anthropic, with side-by-side flagship models, cheapest tiers, and context windows. Pricing data syncs weekly from a continuously-updated model catalog — last updated August 11, 2026.

Who wins on what

Cheapest input tokens

$0.03/1M

OpenAI

gpt-oss-20b — $0.03/1M input

Cheapest output tokens

$0.14/1M

OpenAI

gpt-oss-20b — $0.14/1M output

Longest context window

2.0M

OpenAI

gpt-5.4 (>272K context length) — 2.0M input tokens

Lowest average output cost

$26.11/1M

Anthropic

Provider-wide average across 36 models

Largest model catalog

129 models

OpenAI

More options to match cost vs capability

Cheapest cached input

$0.02/1M

OpenAI

GPT-5.6 Luna — $0.02/1M cached read

Most reasoning models

12 models

OpenAI

Models with dedicated reasoning / thinking support

Most vision models

13 models

OpenAI

Models that accept image input

Side-by-side

129 models

OpenAI

Full OpenAI pricing →

Cheapest input

$0.030

gpt-oss-20b

Cheapest output

$0.140

gpt-oss-20b

Longest context

2.0M

gpt-5.4 (>272K context length)

Avg output / 1M

$29.47

Across catalog

Cheapest cached input

$0.020

GPT-5.6 Luna

Batch discount

50% off output

35 models with published batch rates

ModelIn/1MOut/1MCtx
GPT-5.6
VisionReasoningToolsCache
Batch −50%
$5.00$30.001.1M
GPT-5.6 Sol
VisionReasoningToolsCache
Batch −50%
$5.00$30.001.1M
GPT-5.6 Terra
VisionReasoningToolsCache
Batch −50%
$2.00$12.001.1M
GPT-5.6 Luna
VisionReasoningToolsCache
Batch −50%
$0.200$1.201.1M
GPT-Realtime-2.1
VisionReasoningToolsCache
$4.00$24.00128K
gpt-oss-20b$0.030$0.140131K
36 models

Anthropic

Full Anthropic pricing →

Cheapest input

$0.250

Claude 3 Haiku

Cheapest output

$1.25

Claude 3 Haiku

Longest context

1.0M

Claude 4 Sonnet (2025-05-14)

Avg output / 1M

$26.11

Across catalog

Cheapest cached input

$0.200

Claude Sonnet 5

Batch discount

50% off output

19 models with published batch rates

ModelIn/1MOut/1MCtx
Claude Opus 5
VisionReasoningToolsCache
Batch −50%
$5.00$25.001.0M
Claude Sonnet 5
VisionReasoningToolsCache
Batch −50%
$2.00$10.001.0M
Claude Fable 5
VisionReasoningToolsCache
Batch −50%
$10.00$50.001.0M
Claude Opus 4.8
VisionReasoningToolsCache
Batch −50%
$5.00$25.001.0M
Claude Opus 4.7Batch −50%$5.00$25.00200K
Claude 3 Haiku$0.250$1.25200K

All prices in USD per 1 million tokens. Showing top 6 models per provider, sorted by output cost.

Batch badges mark models with a published asynchronous batch rate. A model without one is billed at its standard rate — our catalog does not carry batch rates for every model, so an absent badge is not evidence that no batch tier exists.

Frequently asked questions

Is OpenAI or Anthropic cheaper?

OpenAI has the cheaper entry point at $0.14/1M output (gpt-oss-20b — $0.14/1M output). Provider-wide, OpenAI averages $29.47/1M output against $26.11/1M for Anthropic. Which is cheaper for you depends on which model tier your workload actually needs.

How much do OpenAI and Anthropic cost per 1M tokens?

OpenAI starts at $0.140 per 1M output tokens (gpt-oss-20b), with its current flagship GPT-5.6 at $30.00. Anthropic starts at $1.25 (Claude 3 Haiku), with Claude Fable 5 at $50.00. Input tokens cost less than output on both.

Which has the larger context window, OpenAI or Anthropic?

OpenAI — gpt-5.4 (>272K context length) — 2.0M input tokens. For comparison, OpenAI's largest is 2.0M tokens (gpt-5.4 (>272K context length)) and Anthropic's is 1.0M tokens (Claude 4 Sonnet (2025-05-14)).

Which has more reasoning models, OpenAI or Anthropic?

OpenAI lists 12 reasoning models and Anthropic lists 4. Reasoning models bill their internal thinking as output tokens, so a reasoning call costs several times a standard completion of the same visible length — compare them on total tokens billed, not headline rate.

Do OpenAI and Anthropic offer batch discounts?

Both publish discounted rates for asynchronous batch jobs: OpenAI on 35 models in this catalog and Anthropic on 19, both at 50% off output tokens. Batch trades real-time responses for a lower rate, so it fits backfills, evaluations and bulk enrichment rather than anything user-facing.

Should I switch from OpenAI to Anthropic to save money?

Only if the cheaper model still meets your quality bar. Token price is one input; the ones that decide your bill are prompt size, response length, retries and how much conversation history you resend each turn. Model the switch against your real traffic before committing. Prices here are current as of August 11, 2026.

Related comparisons

See the full AI model leaderboard — every model ranked by intelligence, real-world usage, and value.

Run the numbers for your workload

Calcaas multiplies per-token costs by your real usage patterns — inputs, outputs, retries, and conversation history — across both providers in one model.