GLM-5

Z.AI's fifth-generation flagship, a 744B-parameter MoE activating 40B per token with DeepSeek Sparse Attention, aimed at agentic engineering and long-horizon work.

Params: 744B, 40B active
Context: 200K tokens
Max output: 131K
Released: Feb 2026
License: MIT

Visit Z.AI Compare vs other LLMs →

Hosted API price

List price via Z.AI · per 1M tokens

USD

Input: $1.00 /1M
Output: $3.20 /1M
Cheapest: $0.60 / $1.60 via Geodd
Hosted by: 6 providers

Hosted API pricing

How we price

Order

Providers are listed cheapest input rate first, then by output rate, a tie going to the creator's own listing. A provider that lists the model without a published rate sits at the end.

Prices

Base-tier, on-demand rates in USD per 1M tokens, converted at ECB reference rates where a provider publishes in another currency. Cached-input and batch rates show only where a provider publishes them. "Cheapest" marks the one provider with the lowest input and output pair; a tie wears no badge.

Cost

Input rate times the input tokens plus output rate times the output tokens, at the monthly volume set above the table. Cached-input, batch, long-context and reasoning-token billing are not modelled.

Transparency and funding

Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.

Every provider serving GLM-5, cheapest input first. Set your monthly volume to see what each would bill.

Input M Output M

Every provider serving GLM-5, with its price per 1M input and output tokens and what the volume above would cost
Provider	Input /1M	Output /1M	Cached input /1M	Cost at 10M in + 2M out	Link
Geodd Cheapest	$0.60	$1.60		$9.20	Visit website
Z.AI Creator	$1.00	$3.20	$0.20	$16.40	Visit website
DigitalOcean	$1.00	$3.20	$0.20	$16.40	Visit website
Lyceum	$1.00	$3.20		$16.40	Visit website
Novita	$1.00	$3.20	$0.20	$16.40	Visit website
Azure	$1.10	$3.52	$0.22	$18.04	Visit website

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

GLM-5 needs about 414 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 8× A100 at about $10,080 a month.

That costs the same as roughly 7B tokens a month on Z.AI's API. Self-hosting is more expensive below that volume.

Single node ≤ 8

What it costs to self-host GLM-5 at each precision: the memory it needs, the cheapest rental that holds it, that rental's month and the API volume it pays off at
Precision	Memory	Cheapest fit (32K context)(Full 200K context)	Cost /mo	Break-even vs API
4-bit INT4 / FP4	414 GB429 GB	8× A100	$10,080$10,080	7B tokens /mo7B tokens /mo
8-bit FP8 / INT8	747 GB755 GB	4× B300	$22,666$22,666	17B tokens /mo17B tokens /mo
16-bit FP16 / BF16	1,493 GB1,508 GB	8× B300	$45,331$45,331	33B tokens /mo33B tokens /mo

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.

Similarly priced models

The models nearest GLM-5 by blended rate, each at its own cheapest provider.

Models priced nearest GLM-5 per 1M tokens, each at its cheapest provider
Model	Blended /1M	Input /1M	Output /1M	Context	Cutoff	vs GLM-5
Gemini 2.5 Flash Google Cloud	$0.6667	$0.30	$2.50	1M	Jan 2025	−13%
Gemini 3.5 Flash-Lite Google Cloud	$0.6667	$0.30	$2.50	1M	Mar 2026	−13%
GPT-3.5 Turbo OpenAI	$0.6667	$0.50	$1.50	16K	Sep 2021	−13%
Mistral Large 3 Mistral	$0.6667	$0.50	$1.50	256K		−13%
Qwen3.5-Plus Alibaba Cloud	$0.7333	$0.40	$2.40	1M		−4%
GLM-5 This model Z.AI	$0.7667	$0.60	$1.60	200K
DeepSeek R1 Distill Llama 70B DeepSeek	$0.80	$0.80	$0.80	131K		+4%
GLM-4.7 Z.AI	$0.8667	$0.60	$2.20	205K		+13%
Gemini 3 Flash Preview Google Cloud	$0.9167	$0.50	$3.00	1M	Jan 2025	+20%
Qwen3.6-Plus Alibaba Cloud	$0.9167	$0.50	$3.00	1M		+20%
GLM-5.2 Z.AI	$0.95	$0.70	$2.20	1M		+24%

USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Runpod

Spin Up a GPU in Seconds - No Setup Required

Deploy AI Models Globally in 30+ Regions

Scale from 0 to 1000 GPUs on Demand

PyTorch, Tensorflow & Docker Ready

Learn more

Capabilities

What GLM-5 accepts and can do, as published by Z.AI.

Text Accepts and generates natural-language text.
Reasoning Thinks before it answers, always on or as a switchable mode.
Function calling Connect to external tools, APIs, and systems.
Structured output Return responses in structured formats like JSON.

Common questions

What is GLM-5 good for?

Complex system engineering and long-range agent work. MIT weights allow commercial self-hosting, on a multi-GPU node.

When is GLM-5 not a good fit?

Text only, and no built-in web search, though a host or your own tool loop can add one. GLM-5.1 has since replaced it, and self-hosting takes a large multi-GPU node.

What is the cheapest way to run GLM-5?

Hosted, unless you push serious volume. Z.AI charges $1.00 in / $3.20 out per 1M tokens. The cheapest rental that fits is 8x A100 at $10,080 a month, which costs the same as about 7B tokens a month on that API.

How does GLM-5 compare with GLM-5.1?

Z.AI says GLM-5.1 codes much better than GLM-5 and, unlike earlier models that run out of ideas early, keeps improving its answer over many rounds of tool calls on long agent tasks. Same text input. GLM-5 lists at 29% less per 1M input tokens and 27% less per 1M output tokens than GLM-5.1. It came out 2 months earlier and has a smaller context (200K against 205K tokens).

Can I self-host GLM-5?

Yes. The weights are MIT licensed. At 4-bit it needs about 414 GB of GPU memory, which starts at roughly $10,080 a month on the cheapest rental that fits.