DEV Community: UplinkSwitch

Claude API pricing, October 2026: every model per million tokens

UplinkSwitch — Sun, 18 Oct 2026 14:00:00 +0000

Input, cached and output prices for every Claude model, Anthropic's list price next to ours, and what a chat turn and a long-document request cost on Opus, Sonnet and Haiku.

How the Claude API is priced

Claude API calls are billed per token, at three rates per million tokens: input, cached input and output. Output costs several times input on every Claude model, and cached input costs a fraction of it. There is no seat or plan price; a request costs its tokens times the rates of the model it named.

Every Claude model

12 Claude models, newest first. Each row shows our price and Anthropic's list price.

Model	Input	Cached	Output	List input	List cached	List output
Claude Fable 5.1	$6.00	$0.15	$30.00	$10.00	$0.25	$50.00
Claude Opus 5.5	$2.10	$0.12	$6.00	$4.00	$0.20	$20.00
Claude Sonnet 5.5	$1.20	$0.12	$3.00	$2.00	$0.20	$10.00
Claude Haiku 5.5	$0.06	$0.0125	$0.15	$0.10	$0.01	$0.50
Claude Fable 5	$3.00	$0.30	$15.00	$10.00	$1.00	$50.00
Claude Opus 5	$1.50	$0.15	$7.50	$5.00	$0.50	$25.00
Claude Sonnet 5	$1.20	$0.12	$3.00	$2.00	$0.20	$10.00
Claude Opus 4.8	$1.50	$0.15	$7.50	$5.00	$0.50	$25.00
Claude Opus 4.7	$1.50	$0.15	$7.50	$5.00	$0.50	$25.00
Claude Opus 4.6	$1.50	$0.15	$7.50	$5.00	$0.50	$25.00
Claude Sonnet 4.6	$1.80	input rate	$9.00	$3.00	input rate	$15.00
Claude Haiku 4.5	$0.60	$0.06	$3.00	$1.00	$0.10	$5.00

USD per million tokens. Ours from the live catalog; list is the price on Anthropic's pricing page, the read date on each model page.

The lowest price is Claude Haiku 5.5: $0.06 in and $0.15 out.

Two requests, worked

A chat turn of 2,000 input and 500 output tokens, and a long-document request of 50,000 input and 2,000 output. Both sizes are assumptions; nothing is counted at the cached rate.

Model	Chat turn, list	Chat turn, here	Long document, list	Long document, here
Claude Opus 5.5	$0.018	$0.0072	$0.24	$0.117
Claude Sonnet 5.5	$0.009	$0.0039	$0.12	$0.066
Claude Haiku 5.5	$0.0005	$0.0002	$0.006	$0.0033

Input tokens times the input rate plus output tokens times the output rate, per million.

Claude Opus 5.5: $2.10 in and $6.00 out here, against $4.00 and $20.00 at list, 47% under on input and 70% on output.

Claude Sonnet 5.5: $1.20 in and $3.00 out here, against $2.00 and $10.00 at list, 40% under on input and 70% on output.

Claude Haiku 5.5: $0.06 in and $0.15 out here, against $0.10 and $0.50 at list, 40% under on input and 70% on output.

Cached input

When a request repeats a prefix the API has seen, such as a long system prompt or a document, the response reports those tokens as cached and they bill at the cached rate in the first table. A model with no cached rate bills them at its input rate.

Calling it

The Anthropic SDK takes a base URL and a key. Every Claude id in the table works with the same two values; the OpenAI SDKs work too.

from anthropic import Anthropic

client = Anthropic(
    base_url="https://api.uplinkswitch.com/v1",
    api_key="sk-uplink-live-...",
)

message = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Summarize this in one line."}],
)
print(message.content[0].text)

More in the SDK guide and the Claude Code guide.

Questions

Is there a subscription? No. Credit is prepaid from $10, and each top-up locks your prices for 30 days.

Which Claude model should I use? The cheapest one that does the task. Haiku for short, high-volume work and subagents, Sonnet for most coding and writing, Opus where the harder answer is worth the price. Run the same prompt on each and compare.

Where is the full price list? On the Claude pricing page, and every other model on the Models page. To see what your own usage would cost, re-price an export on the savings page.

Originally published at uplinkswitch.com. Prices are from our live catalog on 2026-10-17; the original page re-reads them on every load.

SillyTavern with uncensored models on one OpenAI-compatible key

UplinkSwitch — Sun, 18 Oct 2026 06:45:15 +0000

Connect SillyTavern to uncensored builds and DeepSeek through Chat Completion and a Custom (OpenAI-compatible) source: one base URL, one key, one model id.

Where the settings are

Open API Connections in the top bar. Set API to Chat Completion and Chat Completion Source to Custom (OpenAI-compatible).

The connection

Field	Value
Custom Endpoint (Base URL)	`https://api.uplinkswitch.com/v1`
Custom API Key	`sk-uplink-live-...`
Enter a Model ID	`roleplay-uncensored`

Stop the URL at /v1. SillyTavern adds /chat/completions and /models itself. The key comes from the console.

Click Connect. Available Models then lists every model id the key can call, or type one into Enter a Model ID. Any chat model on the Models page works here.

Models for roleplay

Model	Id	How it answers	In / 1M	Out / 1M
Roleplay Uncensored	`roleplay-uncensored`	Straight away, with no thinking step.	$0.30	$0.60
DeepSeek V4 Pro Uncensored	`deepseek-v4-pro-uncensored`	Thinks briefly before it answers.	$0.75	$0.75
GLM-5.3 Abliterated	`glm-5.3-abliterated`	Thinks before it answers. Give it a Max Response Length of 1000 or more.	$2.25	$2.25
DeepSeek V4 Flash 0423	`deepseek-v4-flash`	The standard DeepSeek V4 Flash, not an uncensored build.	$0.0279	$0.0558

Response settings

Max Response Length (tokens) caps each reply, and a model that thinks spends part of it thinking. At 300, a GLM-5.3 Abliterated reply stopped mid-sentence; at 1000 it finished.
Streaming works on all four models above.
Top K at 0 means off, and the API treats it that way. A value above 0 goes to the model.

Each reply is billed by its tokens from the prepaid balance. The console's Usage page lists every request with its charge.

Check it

Confirm the key before you connect:

curl https://api.uplinkswitch.com/v1/models -H "Authorization: Bearer $UPLINKSWITCH_API_KEY"

A list of model ids means the key works. Then click Connect in SillyTavern.

Questions

What base URL does SillyTavern need? https://api.uplinkswitch.com/v1, with Chat Completion Source set to Custom (OpenAI-compatible). SillyTavern adds /chat/completions and /models itself.

Can I use DeepSeek in SillyTavern? Yes. deepseek-v4-flash costs $0.0279 in and $0.0558 out per million tokens. deepseek-v4-pro-uncensored is the uncensored build.

Which uncensored models work in SillyTavern? roleplay-uncensored, deepseek-v4-pro-uncensored and glm-5.3-abliterated. All three stream.

What happens when the balance runs out? Requests return 402 with the code insufficient_balance until you top up.

Originally published at uplinkswitch.com. Prices are from our live catalog on 2026-10-18; the original page re-reads them on every load.

LLM API pricing comparison, October 2026

UplinkSwitch — Sat, 17 Oct 2026 22:27:54 +0000

Input, cached and output prices for every listed chat model with a developer list price, grouped by developer, the developer's price next to ours. Rebuilt from the live catalog.

The short answer

66 chat models from 11 developers have a developer list price on record. 66 of them are priced here under that list price on both input and output, up to 95% under on the widest gap. The lowest-priced model from each developer is below, then every model grouped by developer.

Developer	Lowest-priced model	Input	Output	List
Anthropic	Claude Haiku 5.5	$0.06	$0.15	$0.10 / $0.50
DeepSeek	DeepSeek V4 Flash 0423	$0.0279	$0.0558	$0.30 / $1.20
Google	Gemini 3.1 Flash Lite Preview	$0.15	$0.90	$0.25 / $1.50
MiniMax	MiniMax M2.5	$0.162	$0.648	$0.27 / $1.08
Moonshot AI	Kimi K2.5	$0.27	$1.35	$0.45 / $2.25
OpenAI	GPT-6 Luna	$0.0625	$0.15	$0.10 / $0.50
Qwen	Qwen3.7 Plus	$0.192	$0.768	$0.32 / $1.28
Tencent	Hy3	$0.0417	$0.1668	$0.139 / $0.556
xAI	Grok 4.20 Multi-Agent (0309)	$0.75	$1.50	$1.25 / $2.50
Xiaomi	MiMo V2.6 Flash	$0.084	$0.09	$0.14 / $0.28
Z.ai	GLM 5.3 Flash	$0.084	$0.15	$0.15 / $0.50

Lowest by output price, then input. A model that did not answer the last real-model check (/proof) is skipped.

Anthropic: 12 models

Model	Input	Cached	Output	List input	List output	Under list
Claude Fable 5	$3.00	$0.30	$15.00	$10.00	$50.00	70% in, 70% out
Claude Fable 5.1	$6.00	$0.15	$30.00	$10.00	$50.00	40% in, 40% out
Claude Haiku 4.5	$0.60	$0.06	$3.00	$1.00	$5.00	40% in, 40% out
Claude Haiku 5.5	$0.06	$0.0125	$0.15	$0.10	$0.50	40% in, 70% out
Claude Opus 4.6	$1.50	$0.15	$7.50	$5.00	$25.00	70% in, 70% out
Claude Opus 4.7	$1.50	$0.15	$7.50	$5.00	$25.00	70% in, 70% out
Claude Opus 4.8	$1.50	$0.15	$7.50	$5.00	$25.00	70% in, 70% out
Claude Opus 5	$1.50	$0.15	$7.50	$5.00	$25.00	70% in, 70% out
Claude Opus 5.5	$2.10	$0.12	$6.00	$4.00	$20.00	47% in, 70% out
Claude Sonnet 4.6	$1.80	input rate	$9.00	$3.00	$15.00	40% in, 40% out
Claude Sonnet 5	$1.20	$0.12	$3.00	$2.00	$10.00	40% in, 70% out
Claude Sonnet 5.5	$1.20	$0.12	$3.00	$2.00	$10.00	40% in, 70% out

DeepSeek: 6 models

Model	Input	Cached	Output	List input	List output	Under list
DeepSeek V4 Flash 0423	$0.0279	$0.0056	$0.0558	$0.30	$1.20	90% in, 95% out
DeepSeek V4 Flash 0731	$0.084	$0.0033	$0.09	$0.14	$0.28	40% in, 67% out
DeepSeek V4 Flash Vision Exp	$0.132	$0.0042	$0.396	$0.30	$1.20	56% in, 67% out
DeepSeek V4 Pro 0423	$0.792	$0.0338	$1.92	$1.32	$3.96	40% in, 51% out
DeepSeek V4 Pro 0813	$0.75	$0.0625	$1.188	$1.32	$3.96	43% in, 70% out
DeepSeek V4.1 Flash	$0.06	$0.0063	$0.18	$0.30	$1.20	80% in, 85% out

Google: 11 models

Model	Input	Cached	Output	List input	List output	Under list
Gemini 2.5 Flash	$0.18	$0.018	$1.50	$0.30	$2.50	40% in, 40% out
Gemini 2.5 Pro	$0.75	$0.075	$6.00	$1.25	$10.00	40% in, 40% out
Gemini 3 Flash Preview	$0.30	$0.03	$1.80	$0.50	$3.00	40% in, 40% out
Gemini 3.1 Flash Lite Preview	$0.15	$0.015	$0.90	$0.25	$1.50	40% in, 40% out
Gemini 3.1 Pro Preview	$1.20	$0.12	$7.20	$2.00	$12.00	40% in, 40% out
Gemini 3.5 Flash	$0.90	$0.09	$5.40	$1.50	$9.00	40% in, 40% out
Gemini 3.5 Flash Lite	$0.18	$0.018	$1.50	$0.30	$2.50	40% in, 40% out
Gemini 3.6 Flash	$0.45	$0.045	$2.25	$0.75	$3.75	40% in, 40% out
Gemini 3.7 Flash	$0.45	$0.045	$2.25	$0.75	$3.75	40% in, 40% out
Gemini 3.8 Flash	$0.45	$0.045	$2.25	$0.75	$3.75	40% in, 40% out
Gemini 4 Argon	$0.60	$0.06	$3.00	$2.00	$10.00	70% in, 70% out

MiniMax: 2 models

Model	Input	Cached	Output	List input	List output	Under list
MiniMax M2.5	$0.162	$0.0162	$0.648	$0.27	$1.08	40% in, 40% out
MiniMax M2.7	$0.18	$0.036	$0.72	$0.30	$1.20	40% in, 40% out

Moonshot AI: 4 models

Model	Input	Cached	Output	List input	List output	Under list
Kimi K2.5	$0.27	$0.042	$1.35	$0.45	$2.25	40% in, 40% out
Kimi K2.6	$0.57	$0.096	$2.40	$0.95	$4.00	40% in, 40% out
Kimi K2.7 Code	$0.426	$0.09	$2.10	$0.71	$3.50	40% in, 40% out
Kimi K3	$0.90	$0.09	$4.50	$3.00	$15.00	70% in, 70% out

OpenAI: 13 models

Model	Input	Cached	Output	List input	List output	Under list
GPT-4o	$1.50	input rate	$6.00	$2.50	$10.00	40% in, 40% out
GPT-4o mini	$0.09	input rate	$0.36	$0.15	$0.60	40% in, 40% out
GPT-5.3-Codex	$1.05	$0.105	$8.40	$1.75	$14.00	40% in, 40% out
GPT-5.4	$1.50	$0.15	$9.00	$2.50	$15.00	40% in, 40% out
GPT-5.4 Mini	$0.45	$0.045	$2.70	$0.75	$4.50	40% in, 40% out
GPT-5.5	$1.50	$0.15	$9.00	$5.00	$30.00	70% in, 70% out
GPT-5.6 Luna	$0.12	$0.012	$0.72	$0.20	$1.20	40% in, 40% out
GPT-5.6 Sol	$1.20	$0.12	$3.00	$2.00	$10.00	40% in, 70% out
GPT-5.6 Terra	$1.20	$0.12	$3.60	$2.00	$12.00	40% in, 70% out
GPT-6 Astra	$3.00	$0.30	$15.00	$10.00	$50.00	70% in, 70% out
GPT-6 Luna	$0.0625	$0.0125	$0.15	$0.10	$0.50	37% in, 70% out
GPT-6 Sol	$0.60	$0.12	$3.00	$2.00	$10.00	70% in, 70% out
GPT-6.1 Sol	$0.60	$0.06	$3.00	$2.00	$10.00	70% in, 70% out

Qwen: 3 models

Model	Input	Cached	Output	List input	List output	Under list
Qwen3.6 27B	$0.18	$0.018	$1.20	$0.30	$2.00	40% in, 40% out
Qwen3.7 Max	$0.885	$0.177	$2.655	$1.475	$4.425	40% in, 40% out
Qwen3.7 Plus	$0.192	$0.0384	$0.768	$0.32	$1.28	40% in, 40% out

Tencent: 2 models

Model	Input	Cached	Output	List input	List output	Under list
Hy3	$0.0417	input rate	$0.1668	$0.139	$0.556	70% in, 70% out
Hy4 preview	$0.2502	input rate	$0.75	$0.834	$2.50	70% in, 70% out

xAI: 5 models

Model	Input	Cached	Output	List input	List output	Under list
Grok 4.20 Multi-Agent (0309)	$0.75	$0.12	$1.50	$1.25	$2.50	40% in, 40% out
Grok 4.3	$0.75	$0.12	$1.50	$1.25	$2.50	40% in, 40% out
Grok 4.5	$0.60	$0.09	$1.80	$2.00	$6.00	70% in, 70% out
Grok 4.6	$0.60	$0.15	$1.80	$2.00	$6.00	70% in, 70% out
Grok 4.7	$0.60	$0.15	$1.80	$2.00	$6.00	70% in, 70% out

Xiaomi: 4 models

Model	Input	Cached	Output	List input	List output	Under list
MiMo V2.6 Flash	$0.084	$0.0025	$0.09	$0.14	$0.28	40% in, 67% out
MiMo V2.6 Pro	$0.261	$0.0038	$0.30	$0.435	$0.87	40% in, 65% out
MiMo-V2.5	$0.084	$0.0017	$0.168	$0.14	$0.28	40% in, 40% out
MiMo-V2.5-Pro	$0.261	$0.0022	$0.522	$0.435	$0.87	40% in, 40% out

Z.ai: 4 models

Model	Input	Cached	Output	List input	List output	Under list
GLM 5.1	$0.5796	$0.1076	$1.8216	$0.966	$3.036	40% in, 40% out
GLM 5.2	$0.18	$0.051	$0.60	$0.60	$2.00	70% in, 70% out
GLM 5.3	$0.3276	$0.0608	$1.0296	$1.092	$3.432	70% in, 70% out
GLM 5.3 Flash	$0.084	$0.009	$0.15	$0.15	$0.50	44% in, 70% out

Method

The rows are the price index, built on every request from GET https://api.uplinkswitch.com/v1/catalog. A list price is the developer's own published rate per million tokens, read from its pricing page; the date it was read is on each model page. The percent under list is rounded down to a whole number, and 0% means at or above list on that class.

Left out: models with no list price on record (the Models page lists them with our price alone), and third-party modified builds. Dated snapshots and effort variants fold into their base model.

This post keeps its URL and is refreshed monthly with the month in the title. The numbers above are whatever the catalog says now.

Get the data

The same rows as JSON and CSV, cached for five minutes, each with an x-uplink-generated header carrying the build time. Cite the index, not this post.

Pricing by family, each with the developer's list price next to ours: Claude, GPT, Gemini, Grok, DeepSeek, Kimi, GLM, Qwen.

Originally published at uplinkswitch.com. Prices are from our live catalog on 2026-10-17; the original page re-reads them on every load.

Kimi, GLM and Qwen coding plans vs pay-per-token API, October 2026

UplinkSwitch — Sat, 17 Oct 2026 22:27:12 +0000

What the Kimi, Z.ai GLM and Alibaba Qwen coding plans cost and include, whether DeepSeek has a code plan, and what the same monthly money buys per token.

The plans, October 2026

Kimi, Z.ai and Alibaba sell monthly coding plans. Each caps usage per 5-hour window instead of charging per token. DeepSeek sells no plan.

Plan	Price	Included	Source
Kimi Plus	$19/mo	Kimi Code included, rolling 5-hour limit	www.kimi.ai
Kimi Pro	$39/mo	Kimi Code included, 2x agent credits	www.kimi.ai
Kimi Max	$99/mo	Kimi Code included, 5x agent credits	www.kimi.ai
Kimi Ultra	$199/mo	Kimi Code included, 10x agent credits	www.kimi.ai
Z.ai GLM Coding Plan	from $18/mo	Lite 2,000, Pro 12,000, Max 28,000 credits per 5 hours; GLM-5.3 and GLM-5.3-Flash	docs.z.ai
Alibaba Coding Plan Pro	$50/mo	6,000 requests per 5 hours, 45,000 a week, 90,000 a month	www.alibabacloud.com
DeepSeek none	API only	Pay per token from a prepaid balance	api-docs.deepseek.com

Monthly prices billed monthly, from each developer’s own page, read 2026-10-17. Kimi’s annual billing is lower.

Does DeepSeek have a code plan?

No. DeepSeek's own pricing page lists only pay-per-token API rates, charged against a topped-up balance. Sites selling a "DeepSeek Code Plan" are resellers. Our DeepSeek prices, with DeepSeek's list price next to each, are on the DeepSeek API pricing post.

What a plan’s price buys per token

The same monthly money spent per token here, with no 5-hour window. The mix is an assumption: 10 input tokens for every output token, which is typical of a coding agent re-sending files. Nothing is counted at the cached rate, so cached repeats stretch it further.

Model	Input	Output	$19 buys	$39 buys	$50 buys	$99 buys
Kimi K3	$0.90	$4.50	15M	32M	41M	81M
GLM 5.3	$0.3276	$1.0296	49M	100M	128M	253M
Qwen3.7 Plus	$0.192	$0.768	78M	160M	205M	405M
DeepSeek V4 Pro 0813	$0.75	$1.188	24M	49M	63M	125M
DeepSeek V4.1 Flash	$0.06	$0.18	268M	550M	705M	1,396M

Input and output in USD per million tokens, from the live catalog. "Buys" is millions of tokens at 10:1 input to output.

Plan or per token

A plan fits if you code in one tool every day and run into its window rarely. You pay the same whether you use it or not.

Per token fits uneven use, and work across more than one developer's models. One key reaches Kimi, GLM, Qwen, DeepSeek, Claude, GPT and Gemini, in Cursor, Cline, OpenCode or your own code. There is no window to wait out, and nothing is charged in a month you don't use it.

Questions

Does Kimi K3 work in Cursor or Cline here? Yes. Both take an OpenAI-compatible base URL; the Cursor guide and the Cline guide have the settings. Use kimi-k3 as the model id.

Is there a subscription here? No. Credit is prepaid from $10, and each top-up locks your prices for 30 days.

Where are the full price lists? On the Kimi, GLM, Qwen and DeepSeek pricing pages.

Originally published at uplinkswitch.com. Prices are from our live catalog on 2026-10-17; the original page re-reads them on every load.