How much does GLM 5.2 cost?

On the Respan gateway, input starts at $0.08 per 1M tokens on Relace, and output starts at $2.20 per 1M tokens on Inference.net, after gateway discounts. Averaged across 21 providers, it costs $1.13 per 1M input tokens and $4.89 per 1M output tokens.

Which providers serve GLM 5.2 on Respan?

21 providers serve GLM 5.2: Relace, OpenRouter, Wafer, Inference.net, DeepInfra, MorphLLM, and 13 more. Fireworks, TogetherAI, and Tera require your own provider API key (BYOK).

What is GLM 5.2's context window?

GLM 5.2 accepts up to 1.0M tokens of context on its largest provider route.

How do I use GLM 5.2 with the Respan gateway?

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "zai/glm-5.2". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Over the last 30 days of requests through the Respan gateway, the fastest provider was Baseten at 121 tokens per second, and the lowest response latency was 0.6 seconds on FriendliAI. Speed varies with request size and provider.

Where does GLM 5.2 rank on the Respan Index?

GLM 5.2 scores 40.1 out of 100 on the Respan Index v1.1, ranking #48 of 138 models. By category: Coding 22.7, Agents 57.0, Reasoning 33.6, Knowledge & language 34.3, and Human preference 55.3. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by llmmetric.com.

How does GLM 5.2 score on benchmarks?

GLM 5.2 has 50 published benchmark scores, including Terminal-Bench 2.1 81%, Toolathlon Verified 59.9%, and GPQA Diamond 91.9%. Each score links to where it was published; the data is collected by llmmetric.com.

GLM 5.2

GLM-5.2 is Z.ai's flagship long-horizon reasoning model with a 1M-token context window for agent workflows, coding, and complex multi-step tasks.

Tool Use

# Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://respan.ai/docs/agent-setup.md. Use model "zai/glm-5.2".

Context window: 1.0M tokens
Input from: $0.08/M
Output from: $2.20/M
Providers: 21
Released: Jun 16, 2026

Providers

21 providers serve this model through the Respan gateway, cheapest first. Copy a model ID to call that route directly.

Provider	Context	Input	Output
Relace`relace/z-ai/glm-5.2`	1.0M	$0.08/M	$12.00/M
OpenRouter`openrouter/z-ai/glm-5.2`	1.0M	$0.15/M	$12.00/M
Wafer`wafer/GLM-5.2`	1.0M	$0.23/M	$4.40/M
Inference.net`inference_net/glm-5.2`	1.0M	$0.65/M	$2.20/M
DeepInfra`deepinfra/zai-org/GLM-5.2`	1.0M	$0.75/M	$2.40/M
MorphLLM`morph/morph-glm52-744b`	1M	$1.10/M	$4.10/M
Inceptron`inceptron/zai-org/GLM-5.2`	1.0M	$1.20/M	$4.20/M
NovitaSave 10%`novita/zai-org/glm-5.2`	1.0M	$1.26/M	$3.96/M
IO.net`ionet/zai-org/GLM-5.2`	262K	$1.36/M	$4.36/M
Baseten`baseten/zai-org/GLM-5.2`	1.0M	$1.40/M	$4.40/M
EmpirioLabs`empiriolabs/glm-5-2`	1M	$1.40/M	$4.40/M
Featherless`featherless_ai/zai-org/GLM-5.2`	262K	$1.40/M	$4.40/M
FireworksBYOK only`fireworks_ai/glm-5p2`	1.0M	$1.40/M	$4.40/M
FriendliAI`friendliai/zai-org/GLM-5.2`	1.0M	$1.40/M	$4.40/M
Nebius`nebius/zai-org/GLM-5.2`	1.0M	$1.40/M	$4.40/M
Parasail`parasail/parasail-glm-52`	262K	$1.40/M	$4.40/M
Parasail`parasail/zai-org/GLM-5.2`	262K	$1.40/M	$4.40/M
Parasail`parasail/zai-org/GLM-5.2-FP8`	262K	$1.40/M	$4.40/M
TogetherAIBYOK only`together_ai/zai-org/GLM-5.2`	1.0M	$1.40/M	$4.40/M
Z.ai`zai/glm-5.2`	1M	$1.40/M	$4.40/M
TeraBYOK only`tera/zai-org/GLM-5.2`	262K	$1.49/M	$4.62/M

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

Pricing

Averaged across 21 providers, in USD per 1M tokens after gateway discounts.

Average input: $1.13/M
Average output: $4.89/M
Average cache read: $0.30/M
Average cache write: $0.04/M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$1.49/MTeraLowest$0.08/MRelace

Throughput and latency

30-day averages per provider from requests through the Respan gateway, best first. 11 of 21 providers had traffic in that window.

Output speed

Higher is better

121tokens/sbest, on Baseten

Baseten121 tok/s
FriendliAI97 tok/s
Parasail66 tok/s

Time to first token

Lower is better

2.5son Baseten

Baseten2.5 s

Response latency

Lower is better

0.6sbest, on FriendliAI

FriendliAI0.6 s
Parasail1.0 s
Inceptron1.3 s

Benchmarks

50 published scores for GLM 5.2 from 14 sources. Each score links to where it was published. Percentages are out of 100. Arena ratings are Elo, marked between the lowest and highest rated model on each leaderboard.