BL4ZE Get started

Bring any model.

BL4ZE talks to any OpenAI-compatible endpoint — it doesn't ship with a model, or favor one. Below is standard-tier pricing for 74 chat models, 4 image models and 1 video model across 11 providers, for reference when you're deciding which key to add.

OpenAI (25)

gpt-4.1OpenAI

Input$2.00 /1M

Output$8.00 /1M $0.50 /1M cache-read

gpt-4.1-miniOpenAI

Input$0.40 /1M

Output$1.60 /1M $0.10 /1M cache-read

gpt-4oOpenAI

Input$2.50 /1M

Output$10.00 /1M $1.25 /1M cache-read

gpt-4o-miniOpenAI

Input$0.15 /1M

Output$0.60 /1M $0.075 /1M cache-read

gpt-5-miniOpenAI

Input$0.25 /1M

Output$2.00 /1M $0.025 /1M cache-read

gpt-5.1OpenAI

Input$1.25 /1M

Output$10.00 /1M $0.125 /1M cache-read

gpt-5.2OpenAI

Input$1.75 /1M

Output$14.00 /1M $0.175 /1M cache-read

gpt-5.3-chat-latestOpenAI

Input$1.75 /1M

Output$14.00 /1M $0.175 /1M cache-read

gpt-5.3-codexOpenAI

Input$1.75 /1M

Output$14.00 /1M $0.175 /1M cache-read

gpt-5.4OpenAI

Input$2.50 /1M

Output$15.00 /1M $0.25 /1M cache-read

gpt-5.4-miniOpenAI

Input$0.75 /1M

Output$4.50 /1M $0.075 /1M cache-read

gpt-5.4-nanoOpenAI

Input$0.20 /1M

Output$1.25 /1M $0.02 /1M cache-read

gpt-5.4-proOpenAI

Input$30.00 /1M

Output$180.00 /1M no cache discount published

gpt-5.5OpenAI

Input$5.00 /1M

Output$30.00 /1M $0.50 /1M cache-read

gpt-5.5-openai-compactOpenAI

Input$5.00 /1M

Output$30.00 /1M $0.50 /1M cache-read (gateway alias of gpt-5.5)

gpt-5.5-proOpenAI

Input$30.00 /1M

Output$180.00 /1M no cache discount published

gpt-5.6-lunaOpenAI

Input$0.20 /1M

Output$1.20 /1M $0.02 /1M cache-read

gpt-5.6-solOpenAI

Input$4.00 /1M

Output$20.00 /1M $0.40 /1M cache-read

gpt-5.6-sol-openai-compactOpenAI

Input$4.00 /1M

Output$20.00 /1M $0.40 /1M cache-read (gateway alias of gpt-5.6-sol)

gpt-5.6-terraOpenAI

Input$2.00 /1M

Output$12.00 /1M $0.20 /1M cache-read

gpt-5.6-terra-openai-compactOpenAI

Input$2.00 /1M

Output$12.00 /1M $0.20 /1M cache-read (gateway alias of gpt-5.6-terra)

gpt-6-astraOpenAI

Input$10.00 /1M

Output$50.00 /1M $1.00 /1M cache-read

gpt-oss-120bOpenAI

open-weight — self-hosted or third-party, not billed via the OpenAI API

o1OpenAI

Input$15.00 /1M

Output$60.00 /1M $7.50 /1M cache-read

o3-miniOpenAI

Input$1.10 /1M

Output$4.40 /1M $0.55 /1M cache-read

Anthropic (12)

claude-fable-5Anthropic

Input$10 /1M

Output$50 /1M $1 /1M cache-read

claude-fable-5-1Anthropic

Input$10 /1M

Output$50 /1M $0.25 /1M cache-read

claude-haiku-4-5Anthropic

Input$1 /1M

Output$5 /1M $0.10 /1M cache-read

claude-haiku-4-5-20251001Anthropic

Input$1 /1M

Output$5 /1M $0.10 /1M cache-read

claude-opus-4-5Anthropic

Input$5 /1M

Output$25 /1M $0.50 /1M cache-read

claude-opus-4-6Anthropic

Input$5 /1M

Output$25 /1M $0.50 /1M cache-read

claude-opus-4-7Anthropic

Input$5 /1M

Output$25 /1M $0.50 /1M cache-read

claude-opus-4-8Anthropic

Input$5 /1M

Output$25 /1M $0.50 /1M cache-read

claude-opus-5Anthropic

Input$5 /1M

Output$25 /1M $0.50 /1M cache-read

claude-sonnet-4-5Anthropic

Input$3 /1M

Output$15 /1M $0.30 /1M cache-read

claude-sonnet-4-6Anthropic

Input$3 /1M

Output$15 /1M $0.30 /1M cache-read

claude-sonnet-5Anthropic

Input$2 /1M

Output$10 /1M $0.20 /1M cache-read

Google (13)

gemini-2.5-flashGoogle

Input$0.30 /1M

Output$2.50 /1M $0.03 /1M cache-read

gemini-2.5-flash-liteGoogle

Input$0.10 /1M

Output$0.40 /1M $0.01 /1M cache-read

gemini-2.5-proGoogle

Input$1.25 /1M

Output$10.00 /1M $0.125 /1M cache-read

gemini-3-flash-previewGoogle

Input$0.50 /1M

Output$3.00 /1M $0.05 /1M cache-read

gemini-3-pro-previewGoogle

Input$2.00 /1M

Output$12.00 /1M $0.20 /1M cache-read

gemini-3.1-flash-liteGoogle

Input$0.25 /1M

Output$1.50 /1M $0.025 /1M cache-read

gemini-3.1-flash-lite-previewGoogle

Input$0.25 /1M

Output$1.50 /1M $0.025 /1M cache-read (preview retired 2026-05-25; base rate shown)

gemini-3.1-pro-previewGoogle

Input$2.00 /1M

Output$12.00 /1M $0.20 /1M cache-read

gemini-3.5-flashGoogle

Input$1.50 /1M

Output$9.00 /1M $0.15 /1M cache-read

gemini-3.5-flash-liteGoogle

Input$0.30 /1M

Output$2.50 /1M $0.03 /1M cache-read

gemini-3.6-flashGoogle

Input$0.75 /1M

Output$3.75 /1M $0.075 /1M cache-read

gemini-3.7-flashGoogle

Input$0.75 /1M

Output$3.75 /1M $0.075 /1M cache-read

gemini-3.8-flashGoogle

Input$0.75 /1M

Output$3.75 /1M $0.075 /1M cache-read

DeepSeek (2)

deepseek-v4-flashDeepSeek

Input$0.15 off-peak / 0.30 peak /1M

Output$0.60 off-peak / 1.20 peak /1M · $0.003 off-peak / 0.006 peak/1M cache-read

deepseek-v4-proDeepSeek

Input$0.66 off-peak / 1.32 peak /1M

Output$1.98 off-peak / 3.96 peak /1M · $0.022 off-peak / 0.044 peak/1M cache-read

xAI (9)

grok-4.20-0309-non-reasoningxAI

Input$1.25 /1M

Output$2.50 /1M $0.20 /1M cache-read

grok-4.20-0309-reasoningxAI

Input$1.25 /1M

Output$2.50 /1M $0.20 /1M cache-read

grok-4.20-multi-agentxAI

Input$1.25 /1M

Output$2.50 /1M $0.20 /1M cache-read

grok-4.3xAI

Input$1.25 /1M

Output$2.50 /1M $0.20 /1M cache-read

grok-4.5xAI

Input$2.00 /1M

Output$6.00 /1M $0.30 /1M cache-read

grok-4.5-latestxAI

Input$2.00 /1M

Output$6.00 /1M $0.30 /1M cache-read

grok-4.6xAI

Input$2.00 /1M

Output$6.00 /1M $0.50 /1M cache-read

grok-build-0.1xAI

Input$1.00 /1M

Output$2.00 /1M $0.20 /1M cache-read

grok-composer-2.5-fastCursor Fast tier

Input$3.00 /1M

Output$15.00 /1M $0.50 /1M cache-read (not on xAI’s own pricing page; Cursor’s Fast tier rate shown)

Z.ai (GLM) (3)

glm-5.2Z.ai (GLM)

Input$1.40 /1M

Output$4.40 /1M $0.26 /1M cache-read

glm-5.3Z.ai (GLM)

Input$1.40 /1M

Output$4.40 /1M $0.26 /1M cache-read

glm-5.3-flashZ.ai (GLM)

Input$0.15 /1M

Output$0.50 /1M $0.03 /1M cache-read

Moonshot AI (Kimi) (2)

kimi-k2.7Moonshot AI (Kimi)

Input$0.95 /1M

Output$4.00 /1M $0.19 /1M cache-read

kimi-k3Moonshot AI (Kimi)

Input$3.00 /1M

Output$15.00 /1M $0.30 /1M cache-read

Alibaba (Qwen) (4)

qwen3.7-maxAlibaba (Qwen)

Input$2.50 /1M

Output$7.50 /1M · $0.25 (explicit) / 0.50 (implicit)/1M cache-read

qwen3.7-plusAlibaba (Qwen)

Input$0.40 /1M

Output$1.60 /1M · $0.04 (explicit) / 0.08 (implicit)/1M cache-read

qwen3.8-27bAlibaba (Qwen)

Input$0.50 /1M

Output$3.00 /1M · $0.05 (explicit) / 0.10 (implicit)/1M cache-read

qwen3.8-maxAlibaba (Qwen)

Input$2.00 /1M

Output$6.00 /1M · $0.17 (explicit) / 0.25 (implicit)/1M cache-read

MiniMax (1)

minimax-m3MiniMax

Input$0.30 /1M

Output$1.20 /1M $0.06 /1M cache-read

Xiaomi (MiMo) (1)

mimo-v2.5Xiaomi (MiMo)

Input$0.14 /1M

Output$0.28 /1M $0.0028 /1M cache-read

Cursor (2)

codex-auto-reviewCursor

Cursor/Codex-internal review pass, not sold as a standalone API model

composer-2.5Cursor · two tiers

Input$0.50 /1M$3.00 on Fast

Output$2.50 /1M$15.00 on Fast$0.20 cache-read · $0.50 on Fast

Image generation — billed per image, not per token

gemini-3-pro-imageGoogle

$0.134per image 1K / 2K$0.24 at 4K

gemini-3.1-flash-imageGoogle

$0.067per image $0.045 at 0.5K1K$0.101 at 2K$0.151 at 4K

gpt-image-2OpenAI

$0.211per image 1024² high qualitybilled as tokens, not a flat fee — $15/1M image output

nano-banana-proGoogle

$0.134per image 1K / 2K$0.24 at 4Ksame model as gemini-3-pro-image

Video generation — billed per second of output, not per token

veo-3.1-fastGoogle

$0.10per second 720p with audio$0.12 at 1080p$0.30 at 4K

Researched 2026-09-17, standard/paid-tier rates from each vendor's own docs where published. Not BL4ZE's or BlazeAPI's price — this is what the vendor itself charges, before anything either of those adds for routing or margin. A few things worth knowing before comparing rows directly:

Several providers charge more above a context-length cutoff — often 200K tokens — so the rate shown is the short-context one.
DeepSeek and Qwen publish separate peak/off-peak or region rates; off-peak / Singapore International is shown here.
Cache-read pricing varies a lot by vendor (Anthropic ≈0.1× input, Gemini ≈0.1×, DeepSeek ≈0.01–0.02×) — don't assume one ratio applies everywhere.
Two models aren't independently priced: codex-auto-review is a Cursor/Codex-internal pass, and gpt-oss-120b is open-weight and billed by whoever hosts it, not by OpenAI.

Already have a key

Add it in onboarding

Point bl4ze provider add at any of the above, or pick "custom provider" the first time BL4ZE opens. You pay the vendor directly, at the rates shown.

Install the CLI →

Don't have one yet

Sign in with BlazeAPI

Skips the key entirely — every model your BL4ZE account can reach becomes available the moment you link. 200,000 tokens a day free, pay-as-you-go beyond that.

Get started