GLM-5.1 API pricing

4 providers serve GLM-5.1. GLM-5.1 is an open-weight model (MIT) with 744B total parameters, 40B active per token (mixture-of-experts), up to 203K context. Prices below are per 1M tokens, cheapest input first (self-hosting math further down).

Refreshed 2026-08-23 · list prices from provider APIs

Input floor $1.05/1M at OpenRouter Output floor $3.50/1M at OpenRouter Find your workload winner →

Provider	Input $/1M	Output $/1M	Cache read $/1M	Context
OpenRouter	$1.05	$3.50output floor	$0.525	203K	Use
Alibaba Model Studio	$1.40	$4.40	$0.260	203K	Use
Fireworks AI	$1.40	$4.40	$0.260	203K	Use
Z.ai	$1.40	$4.40	$0.260	200K	Use

The cheapest way to run GLM-5.1

On an illustrative 70M-input / 30M-output monthly workload, today's lowest listed cost is roughly $179/month, at OpenRouter. Input-heavy, output-heavy and cache-heavy workloads can produce different winners. Use the live mix calculator below rather than combining floors from two different providers. On output-token cost alone, a busy self-hosted deployment can work out cheaper (~$1.65/1M output on 6× MI300X). See the math below. Run your own numbers in the breakeven calculator, or read the breakeven math.

Your actual monthly cost

Uses today's list prices

Workload preset Input tokens / month (millions) Output tokens / month (millions) Input served from cache %

Cache read uses the listed cache rate where available; otherwise it falls back to normal input price. Batch, write-cache, volume and negotiated discounts are excluded.

Or run it yourself: GLM-5.1 self-hosting economics

GLM-5.1 is open-weight (MIT), so the API price above competes with the GPU-hour market. 744B total parameters (MoE, ~40B active per token) needs roughly 1024 GB VRAM at FP8 or 512 GB at INT4, KV-cache headroom included.

No single GPU fits GLM-5.1 at FP8 (1024 GB needed), so it needs a multi-GPU node. Cheapest tracked option: 6× MI300X (192 GB each) at roughly $3.00/hr total (RunPod). Rent 6× MI300X at RunPod → INT4 quantization drops the requirement to 512 GB. Tune it in the calculator.

Breakeven estimate: a well-batched vLLM deployment on 6× MI300X ($3.00/hr) at ~1013 aggregate tok/s and 50% utilization works out to roughly $1.65 per 1M output tokens, versus $3.50 via the cheapest output-token API (OpenRouter). Self-hosting wins on cost if you can keep the GPU busy. Planning estimate: throughput varies with hardware, quantization, batch size and context. Tune it in the calculator.

Related models

Model	Cheapest in $/1M	Cheapest out $/1M	Providers
GLM-4.7	$0.400	$1.50	10
GLM-4.6	$0.400	$1.75	8
GLM-4.5 Air	$0.125	$0.450	7
GLM-5.2	$0.610	$1.98	6
GLM-4.5	$0.400	$1.60	6
GLM-5	$0.800	$2.56	5

FAQ

What is the cheapest API for GLM-5.1?

As of 2026-08-23, the lowest input price for GLM-5.1 is OpenRouter at $1.05 per 1M input tokens. The lowest output price is OpenRouter at $3.50 per 1M output tokens. The cheapest provider for a real workload depends on its input/output mix. The next-lowest input price is Alibaba Model Studio at $1.40, a 33% difference.

How much VRAM do you need to self-host GLM-5.1?

GLM-5.1 has 744B parameters, so plan for roughly 1024 GB of VRAM at FP8 or 512 GB at INT4/AWQ, KV-cache headroom included. That exceeds a single GPU. A typical node is 6× MI300X.

← All models · GPU rental prices · Breakeven calculator · Get this data as JSON