GLM 4.7 Flash is a text-only model optimized for high-throughput, low-cost workloads with long context windows. Reach for it when you need fast, reliable processing of large documents or bulk chat tasks without paying premium rates.
Modalities
Text → Text
Below release
fp8
Weights
Published
zai-org/GLM-4.7-Flash
Where it sits in the catalogue Kyma measures
Every number against every text model Kyma prices per token — a stated rule, not a chosen line-up.
$0.081
output $0.54 · cached $0.0135
median $0.977 · best $0.0494
$0.54
per 1M tokens generated
median $3.531 · best $0.2164
95.69%
627 observations since 2026-05-02
median 96.2% · best 100.0%
61.2 tok/s
probe median, one fixed prompt
median 40.8 tok/s · best 308.3 tok/s
4.55 s
probe median, to a complete answer
median 2.50 s · best 0.52 s
203K
max output 65,536
published by Zhipu AI, not measured here
Tick above each rail is this model, below it the other 69. Dashed rule is the field median, solid is its best. Better is left; the axis stops at the 90th percentile, so a few models sit past its right edge.
Usage
How much this model is actually called here.
Rank
#12
of 113 active models
Tokens served
52.9M
all-time
Platform share
1.6%
of all tokens
How it behaves under real clients
The same model answering different prompt shapes — measured, not benchmarked.
| Client | Requests | Tokens | Throughput | To first token | p95 total | Completed |
|---|---|---|---|---|---|---|
| Python App | 156 | 2.0M | 65 tok/s | 111 s | 679 s | 100% |
Grouped by the client that sent the request. Not a benchmark: the same model on different prompt shapes. Throughput and time-to-first-token are medians; p95 is the slowest response in twenty.
Two clocks, and why they disagree
Kyma measures this model twice. Both are real; they answer different questions.
Probe · every 6h · 30 days
4.55s to answer
One fixed prompt, on a schedule, to every model. Comparable, because the model is the only thing that changes.
- Observations
- 627
- Answered by a substitute
- 12
Real traffic · last 7 days
73.93s to answer
Your requests, at the lengths clients actually send. Not comparable between models, but it is what running this one feels like.
- Requests
- 41
- Completed
- 100%
- p95
- 1674.9 s
Caching, as realised · last 7 days
12.2% of input cached
The share of input that actually hit cache, so the effective rate below is what was charged, not a best case.
- Cached input
- 9.1K
- Fresh input
- 65.6K
- List input
- $0.081 /1M
- Effective input
- $0.07278 /1M
The gap is prompt length, not the model degrading. Use the probe figure to choose between models, the traffic figure to budget for your own.
Pricing
Pay per token. Cached input bills at this model’s own cached rate, listed below.
+ Estimate your workload− Estimate your workload
Estimated monthly cost
$11.75
$0.3915 / day on GLM 4.7 Flash
Same workload on:
Estimates use list pricing with cached input at this model's own cached rate. Actual bills depend on real token counts, and every response includes its exact cost.
When to use GLM 4.7 Flash
Where this model earns its cost — and where it doesn't.
Created by Zhipu, GLM 4.7 Flash is an efficient language model that handles up to 203,000 input tokens and generates up to 65,536 output tokens. It operates in the cheap cost and fast speed tiers on Kyma.
The model supports tool calling, reasoning, and structured outputs, and runs on Kyma's OpenAI-compatible API. Prompt caching is available, billing repeated prefixes at this model's cached input rate. Every response includes exact token costs in the usage.cost field, and the X-Kyma-Model header confirms the active model.
It processes text only and does not support vision or multimodal inputs. While optimized for speed and cost efficiency, it sits in the efficient quality tier, so highly complex analytical tasks may require a higher-tier alternative.
Bulk Document Processing
Parse and summarize large text files quickly at low cost.
High-Volume Chat Routing
Handle thousands of concurrent conversational requests with fast response times.
Automated Tool Execution
Run structured workflows and function calling across long conversation histories.
Log Analysis Pipelines
Ingest and categorize extensive text logs using structured JSON outputs.
Not ideal for: Do not use this model for tasks requiring image understanding, advanced creative writing, or highly complex multi-step reasoning where maximum accuracy is critical.
How it compares
Against the peers people actually weigh it against.
Quick start
Up and running in under two minutes.
- 1
Create an API key
Sign up and grab a key from the dashboard — $0.50 free credit, no card required.
Get API key → - 2
Make your first request
Drop in your key and send a chat completion — fully OpenAI-compatible.
curl https://kymaapi.com/v1/chat/completions \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "glm-4.7-flash", "messages": [ {"role": "user", "content": "Explain prompt caching in one paragraph."} ] }' - 3
Stream responses
Add
"stream": trueto receive tokens as they arrive.curl https://kymaapi.com/v1/chat/completions \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "glm-4.7-flash", "stream": true, "messages": [{"role": "user", "content": "Hello!"}] }'
FAQ
Common questions about this model.
What is the context window of GLM 4.7 Flash?
How much does the GLM 4.7 Flash API cost?
Does GLM 4.7 Flash support function calling?
Are the weights for GLM 4.7 Flash publicly available?
Is GLM 4.7 Flash ever served below the precision its creator released it at?
How do I use GLM 4.7 Flash?
Does this model support image inputs?
How does prompt caching affect my costs?
What happens if the serving path degrades during a request?
More models by Zhipu AI
| Model | Context | Input | Output |
|---|---|---|---|
 GLM 5.3 Flash | 1M | $0.101 | $0.338 |
| GLM 5.3 | 1M | $1.89 | $5.94 |
| GLM 5.2 | 1M | $1.101 | $3.76 |
| GLM 5.1 | 203K | $1.89 | $5.94 |
| GLM 4.5 Air | 131K | $0.1945 | $1.272 |