Skip to main content

GLM 5.3 Flash

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks.

Model card on Hugging Face
Input / output price
$0.07 / $0.15 per 1M
Context
1M tokens
Released
Aug 26, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "zai/glm-5.3-flash".

Read docs

Providers

28 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Concentrateconcentrate/glm-5.3-flash
$0.05/M$0.20/M
Waferwafer/glm-5.3-flash
$0.05/M$0.20/M
Pareto Inferenceparetoinference/glm-5.3-flash
$0.06/M$0.20/M
Isoquantisoquant/glm-5.3-flash
$0.07/M$0.15/M
Gonka24gonka24/glm-5.3-flash
$0.07/M$0.19/M

Is a provider down? Fireworks AI status, Mistral AI status, Perplexity status, OpenRouter status

Pricing

Averaged across 28 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.13 per 1M
Average output
$0.44 per 1M
Average cache read
$0.03 per 1M

Cost by volume

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

Highest$0.20/MMorphLLMLowest$0.05/MConcentrate and Wafer

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

66tokens/sacross providers

Faster than 71% of the other 581 models on the gateway

Time to first token

Lower is better

1.1sacross providers

Faster than 66% of the other 144 models on the gateway

Response latency

Lower is better

8.3sacross providers

Faster than 19% of the other 726 models on the gateway

Benchmarks

The Respan Index plus 37 published scores from 9 sources, each linked to where it was published.

45.4/ 100

Respan Index#51 of 172

172 ranked models. The shaded band is this model's 95% interval, 40.1 to 51.5. Gemini 3.6 Flash and DeepSeek V4 Flash fall inside it, so the gap is within the margin of error.
GLM 5.3 Flash by category: Coding 39.6, Agents 61.7, Reasoning 33.5, Knowledge & language 32.0, Human preference 56.6
  1. 51GLM-5.3-Flash45.4
  2. 50Gemini 3.6 Flash45.6
  3. 51DeepSeek V4 Flash45.4

Published scores

BenchmarkScore
Terminal-Bench 2.1autoclaw.z.ai84.3%
LiveBench CodingLiveBench79%
DeepSWE v1.1Datacurve63.4%
DeepSWE v1.1autoclaw.z.ai63.4%
NL2Repoautoclaw.z.ai56.3%
Terminal-Bench 4.0Vals AI37.4%
FrontierSWE V2Proximal Labs18.1%

Apps

Use GLM 5.3 Flash in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

GLM 5.3 Flash FAQ

Short answers, built from the data on this page.

GLM 5.3 Flash is a model from Z.ai, released on Aug 26, 2026. GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks.

On the Respan gateway, input starts at $0.05 per 1M tokens on Concentrate and Wafer, and output starts at $0.15 per 1M tokens on Isoquant, after gateway discounts. Averaged across its 28 providers, it costs $0.13 per 1M input tokens and $0.44 per 1M output tokens.

GLM 5.3 Flash is available through 28 providers: Concentrate, Wafer, Pareto Inference, Isoquant, Gonka24, EmpirioLabs, and 22 more.

GLM 5.3 Flash accepts up to 1M tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "zai/glm-5.3-flash". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

GLM 5.3 Flash scores 45.4 out of 100 on the Respan Index v1.1, ranking #51 of 172 models. By category: Coding 39.6, Agents 61.7, Reasoning 33.5, Knowledge & language 32.0, and Human preference 56.6. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

GLM 5.3 Flash has 37 published benchmark scores, including Terminal-Bench 2.1 84.3%, Toolathlon Verified 78.4%, and GPQA Diamond 90.2%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

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

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.