---
title: "LLM API price changes: a dated log | BroadRouter"
description: "Every LLM API price change we have recorded between checks: prices read from providers' own pages and BroadNet's rate card, with the date, old price and new price."
url: https://broadrouter.com/llm-api-pricing/changes/
lang: en
updated: 2026-11-16
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# LLM API price changes

Updated 2026-11-16

No provider has cut or raised an LLM API price between our saved checks since checks began on 2026-11-14: 0 cuts and 0 rises. 15 data corrections are listed separately and not counted. Each re-check compares the 41 provider price files and BroadNet's rate card with the copy before it.

Each row is one price that differed between two saved copies of a price file. "Seen" is the time the new copy was saved, not the time the provider changed the price. Percentages compare the new price with the old one. Follow new changes with the [RSS feed](https://broadrouter.com/llm-api-pricing/changes.xml).

## When were these prices last re-checked?

Automated re-check 2026-11-16: a weekly job re-fetched 364 provider source pages and looked for each of 1,675 recorded prices. 1,657 still matched the page; 18 are on pages that load prices with JavaScript and were last read with a page renderer, on the date shown in their row. None was flagged. This is an automated read, not a human review, and a price change is added to this log only after a re-read confirms it.

**No price changes yet.** Rows appear here once a re-check finds a provider price that differs from the saved one.

## Data corrections (not price changes)

15 fixes to our own records: a value we had not recorded before, or an edit within 24 hours of a price file's first version. The provider did not change these prices, so they are not counted above.

Data corrections to the price files, newest first
| Saved | Model | Provider | Correction | Before | After |
| --- | --- | --- | --- | --- | --- |
| 2026-11-14 21:47 UTC | [DeepSeek V4 Flash 0731](https://broadrouter.com/llm-api-pricing/deepseek-v4-flash-0731/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.044 |
| 2026-11-14 21:47 UTC | [DeepSeek V4 Pro 0813](https://broadrouter.com/llm-api-pricing/deepseek-v4-pro-0813/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.132 |
| 2026-11-14 21:47 UTC | [GLM 5.1](https://broadrouter.com/llm-api-pricing/glm-5.1/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.26 |
| 2026-11-14 21:47 UTC | [GLM 5.2](https://broadrouter.com/llm-api-pricing/glm-5.2/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.28 |
| 2026-11-14 21:47 UTC | [GLM 5.3](https://broadrouter.com/llm-api-pricing/glm-5.3/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.28 |
| 2026-11-14 21:47 UTC | GLM 5.3 Prime | [Alibaba Cloud Model Studio (fast)](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.56 |
| 2026-11-14 21:47 UTC | [Qwen-Flash](https://broadrouter.com/llm-api-pricing/qwen-flash/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.01 |
| 2026-11-14 21:47 UTC | [Qwen-Max](https://broadrouter.com/llm-api-pricing/qwen-max/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.32 |
| 2026-11-14 21:47 UTC | [Qwen-Plus](https://broadrouter.com/llm-api-pricing/qwen-plus/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.08 |
| 2026-11-14 21:47 UTC | [Qwen3 Coder Flash](https://broadrouter.com/llm-api-pricing/qwen3-coder-flash/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.06 |
| 2026-11-14 21:47 UTC | [Qwen3 VL Plus](https://broadrouter.com/llm-api-pricing/qwen3-vl-plus/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.04 |
| 2026-11-14 21:47 UTC | [Qwen3.7 Flash](https://broadrouter.com/llm-api-pricing/qwen3.7-flash/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.006 |
| 2026-11-14 21:47 UTC | [Qwen3.7 Max](https://broadrouter.com/llm-api-pricing/qwen3.7-max/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.50 |
| 2026-11-14 21:47 UTC | [Qwen3.7 Plus](https://broadrouter.com/llm-api-pricing/qwen3.7-plus/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.08 |
| 2026-11-14 21:47 UTC | [Qwen3.8 27B](https://broadrouter.com/llm-api-pricing/qwen3.8-27b/) | [Alibaba Cloud Model Studio](https://www.alibabacloud.com/help/en/model-studio/model-pricing) | cached input newly recorded | not recorded | USD 0.10 |

Changes seen in archived pricing pages month by month are summarised in [LLM API price changes in October 2026](https://broadrouter.com/blog/llm-api-price-changes-october-2026/).

## How is this log built?

The site keeps one price file per provider in git, 41 in all, each price read from that provider's own pricing page with its source link, the date checked and a quote. BroadNet Inference's rate card is kept the same way. At build time the site reads the saved copies in order and lists every input, cached-input, output, per-minute and plan price that differs from the copy before it. A provider file's first copy is not counted as a change, and corrections to our own records are listed apart from real cuts and rises. For current prices and the cheapest provider per model, see [LLM API pricing by model](https://broadrouter.com/llm-api-pricing/), [the cheapest LLM APIs right now](https://broadrouter.com/llm-api-pricing/cheapest/) and the open [LLM API price index](https://broadrouter.com/data/llm-api-price-index/).

## Questions

### How often are LLM API prices checked?

BroadNet's own rate card (pricing.json) is checked every hour and the site is rebuilt when it changes. Other providers' prices are read from each provider's own pricing page; every row carries the date it was read, and a weekly job re-fetches each source page and flags any price or quote it no longer finds, for re-reading. The newest provider check is 2026-11-15.

### Where does this price history come from?

When a re-check changes a price, the updated provider price file is committed to the site's git repository (one file per provider, 1,671 prices today), and BroadNet's rate card is saved the same way. At build time the site compares each saved copy with the one before it. The newest 120 copies of each file are compared.

### What do "new price listed" and "price removed" mean?

A provider started or stopped publishing a price for that model and tier on its own pricing page between two checks. The row shows the input price per 1M tokens. A removed price can come back later.

### What is a data correction?

A fix to our own records, not a change by the provider: a price we had not recorded (for example a cached-input price added to an existing row), or any edit made within 24 hours of a price file's first version, while the first check was still being finished. Corrections are listed separately and never counted as cuts or rises.

### Are these prices what I pay a provider directly?

Yes: provider rows are read from the provider's own pricing page, not from a router or reseller, and BroadNet rows are BroadNet Inference's own prices. Taxes, volume discounts and regional surcharges are not included; the row notes on each [model page](https://broadrouter.com/llm-api-pricing/) say which region and tier a price applies to.

[Report a wrong price or number](mailto:info@broadnet.ai?subject=Wrong%20price%20or%20number%20on%20the%20LLM%20API%20price%20change%20log&body=Page%3A%20https%3A%2F%2Fbroadrouter.com%2Fllm-api-pricing%2Fchanges%2F%0A%0AWhat%20looks%20wrong%20(model%2C%20figure%20shown%2C%20correct%20figure%20and%20a%20source%20link%20if%20you%20have%20one)%3A%0A) (email to info@broadnet.ai, with this page's link filled in).

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=llm-api-pricing-changes&utm_content=page-cta) ·

**Sources**

-   [AI21 Labs: pricing (checked 2026-11-15)](https://www.ai21.com/pricing/)
-   [AkashML: pricing (checked 2026-11-15)](https://akashml.com/#pricing)
-   [Alibaba Cloud Model Studio: pricing (checked 2026-11-14)](https://www.alibabacloud.com/help/en/model-studio/model-pricing)
-   [Amazon Bedrock: pricing (checked 2026-11-14)](https://aws.amazon.com/bedrock/pricing/)
-   [Amazon Nova (Bedrock): pricing (checked 2026-11-14)](https://aws.amazon.com/bedrock/pricing/)
-   [Anthropic: pricing (checked 2026-11-14)](https://platform.claude.com/docs/en/about-claude/pricing)
-   [Baidu: pricing (checked 2026-11-15)](https://intl.cloud.baidu.com/en/doc/qianfan/s/Jm8r1826a-intl-en)
-   [Baseten: pricing (checked 2026-11-14)](https://www.baseten.co/pricing/)
-   [Cerebras: pricing (checked 2026-11-14)](https://www.cerebras.ai/pricing)
-   [Chutes: pricing (checked 2026-11-15)](https://chutes.ai/pricing)
-   [Cloudflare Workers AI: pricing (checked 2026-11-14)](https://developers.cloudflare.com/workers-ai/platform/pricing/)
-   [Cohere: pricing (checked 2026-11-14)](https://cohere.com/pricing)
-   [CoreWeave Forge (formerly W&B Inference): pricing (checked 2026-11-15)](https://coreweave.com/forge-pricing/inference-agents)
-   [Crusoe Cloud: pricing (checked 2026-11-15)](https://www.crusoe.ai/cloud/pricing#Serverless-Inference)
-   [DeepInfra: pricing (checked 2026-11-14)](https://deepinfra.com/pricing)
-   [DeepSeek: pricing (checked 2026-11-14)](https://api-docs.deepseek.com/quick_start/pricing)
-   [DigitalOcean Gradient (Serverless Inference): pricing (checked 2026-11-14)](https://docs.digitalocean.com/products/inference/details/pricing/)
-   [Featherless: pricing (checked 2026-11-15)](https://featherless.ai/pricing)
-   [Fireworks AI: pricing (checked 2026-11-14)](https://fireworks.ai/pricing)
-   [FriendliAI: pricing (checked 2026-11-15)](https://friendli.ai/pricing)
-   [GMI Cloud: pricing (checked 2026-11-15)](https://www.gmicloud.ai/en/models/maas)
-   [Google (Gemini API): pricing (checked 2026-11-14)](https://aistudio.google.com/docs/pricing)
-   [Google Cloud Vertex AI: pricing (checked 2026-11-14)](https://cloud.google.com/model-garden/pricing)
-   [Groq: pricing (checked 2026-11-14)](https://console.groq.com/docs/models)
-   [Hyperbolic: pricing (checked 2026-11-14)](https://www.hyperbolic.ai/inference)
-   [IBM watsonx.ai: pricing (checked 2026-11-15)](https://www.ibm.com/products/watsonx-ai/pricing)
-   [Microsoft Azure (Azure OpenAI + Foundry Models): pricing (checked 2026-11-14)](https://azure.microsoft.com/en-us/pricing/details/azure-openai/)
-   [MiniMax: pricing (checked 2026-11-14)](https://platform.minimax.io/docs/pricing/overview)
-   [Mistral AI: pricing (checked 2026-11-14)](https://docs.mistral.ai/inference/pricing)
-   [Moonshot AI (Kimi): pricing (checked 2026-11-14)](https://platform.kimi.ai/docs/pricing/chat)
-   [Novita AI: pricing (checked 2026-11-14)](https://novita.ai/pricing)
-   [Nscale: pricing (checked 2026-11-15)](https://docs.nscale.com/docs/ai-services/models)
-   [OpenAI: pricing (checked 2026-11-14)](https://developers.openai.com/api/docs/pricing)
-   [Parasail: pricing (checked 2026-11-14)](https://www.parasail.io/pricing)
-   [Perplexity: pricing (checked 2026-11-15)](https://docs.perplexity.ai/docs/getting-started/pricing)
-   [SambaNova Cloud: pricing (checked 2026-11-14)](https://cloud.sambanova.ai/plans/pricing)
-   [SiliconFlow: pricing (checked 2026-11-14)](https://www.siliconflow.com/pricing)
-   [Together AI: pricing (checked 2026-11-14)](https://www.together.ai/pricing)
-   [Venice: pricing (checked 2026-11-15)](https://docs.venice.ai/pricing)
-   [xAI: pricing (checked 2026-11-14)](https://docs.x.ai/developers/pricing)
-   [Z.ai: pricing (checked 2026-11-14)](https://docs.z.ai/guides/overview/pricing)
-   [BroadNet Inference rate card, pricing.json](https://inference.broadnet.ai/pricing.json?utm_source=broadrouter&utm_medium=referral&utm_campaign=llm-api-pricing-changes&utm_content=sources)
