LLMHOSTING.AI the compute price terminal

GLM-4.5 Air API pricing

7 providers serve GLM-4.5 Air. GLM-4.5 Air is an open-weight model (MIT) with 106B total parameters, 12B active per token (mixture-of-experts), up to 131K 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 $0.125/1M at Pinstripes Output floor $0.450/1M at Pinstripes Find your workload winner →
Provider Input $/1M Output $/1M Cache read $/1M Context
Pinstripes $0.125 $0.450output floor - 128K
Novita AI $0.130 $0.850 - 131K Use
OpenRouter marketplace quote $0.130 $0.850 $0.025 131K Use
Together AI $0.200 $1.10 - 128K Use
Vercel AI Gateway $0.200 $1.10 - 128K Use
Z.ai $0.200 $1.10 - 128K Use
Fireworks AI $0.220 $0.880 - 128K Use

The cheapest way to run GLM-4.5 Air

On an illustrative 70M-input / 30M-output monthly workload, today's lowest listed cost is roughly $22.25/month, at Pinstripes. 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 (~$0.082/1M output on 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

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-4.5 Air self-hosting economics

GLM-4.5 Air is open-weight (MIT), so the API price above competes with the GPU-hour market. 106B total parameters (MoE, ~12B active per token) needs roughly 146 GB VRAM at FP8 or 73 GB at INT4, KV-cache headroom included.

GPU that fits (FP8, single card)VRAMFrom $/hrCheapest at
MI300X 192 GB $0.500community RunPod Rent
B200 180 GB $3.06spot Verda
B300 288 GB $3.75spot Verda
GB300 288 GB $4.31spot Verda
Breakeven estimate: a well-batched vLLM deployment on one MI300X ($0.500/hr) at ~3375 aggregate tok/s and 50% utilization works out to roughly $0.082 per 1M output tokens, versus $0.450 via the cheapest output-token API (Pinstripes). 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

ModelCheapest in $/1MCheapest out $/1MProviders
GLM-4.7 $0.400 $1.50 10
GLM-4.6 $0.400 $1.75 8
GLM-5.2 $0.610 $1.98 6
GLM-4.5 $0.400 $1.60 6
GLM-5 $0.800 $2.56 5
GLM-5.1 $1.05 $3.50 4

FAQ

What is the cheapest API for GLM-4.5 Air?

As of 2026-08-23, the lowest input price for GLM-4.5 Air is Pinstripes at $0.125 per 1M input tokens. The lowest output price is Pinstripes at $0.450 per 1M output tokens. The cheapest provider for a real workload depends on its input/output mix. The next-lowest input price is Novita AI at $0.130, a 4% difference.

How much VRAM do you need to self-host GLM-4.5 Air?

GLM-4.5 Air has 106B parameters, so plan for roughly 146 GB of VRAM at FP8 or 73 GB at INT4/AWQ, KV-cache headroom included. A single MI300X (192 GB) fits it.

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