Fireworks AI · automatic caching
GLM 5.1
Z.ai open-weights model with strong agentic coding behaviour and balanced input/output pricing that favours output-heavy generation.
Specifications
At a glance.
| Provider | Fireworks AI |
| Family | glm_moe_dsa |
| Released | 2026-04 |
| License | Open weights |
| Context window | 203K tokens |
| Max output | 131K tokens |
| Parameters | 744B |
| Modalities | text |
| Tool calling | Yes |
| Reasoning mode | Yes |
| Caching | automatic |
| Batch discount | No batch tier |
Measured by Zumik
What reuse looks like here.
What you actually pay once caching works.
At a typical 55% prefix reuse, a million input tokens on GLM 5.1 effectively costs $0.77 instead of $1.40 - blending to roughly $1.68 with a 25% output share. There is no batch tier, so cost control here leans on caching and routing.
Estimate it for your workloadRoutes through these aliases:
Call it
Same OpenAI client, this model.
from openai import OpenAI
client = OpenAI(base_url="https://api.zumik.ai/v1", api_key="zk_live_...")
r = client.responses.create(
model="glm-5p1", # or an alias like code.balanced
input="Draft a fix for the failing test.",
)
print(r.usage.input_tokens_cached) # confirm reuseCompare
How GLM 5.1 stacks up.
Similar models
Other options for these workloads.
Frequently asked
GLM 5.1, answered.
How much does GLM 5.1 cost?
GLM 5.1 is $1.40 per million input tokens and $4.40 per million output tokens through Zumik. Cache reads are $0.26 per million, a 81% discount on input.
What is GLM 5.1's context window?
GLM 5.1 supports a 203K-token context window with up to 131K output tokens.
Does GLM 5.1 support prompt caching?
Yes. Fireworks AI uses Automatic prompt caching (serverless and dedicated) caching. In the Zumik corpus, GLM 5.1 shows a median cache capture of 83% on agent workloads.
Which Zumik aliases route to GLM 5.1?
GLM 5.1 is a candidate for the code.balanced alias, selected when it wins under current routing policy.
Run GLM 5.1 with reuse measured.
Point an OpenAI client at Zumik and see exactly how much of this model's input you are reusing.