- 1M context
- Reasoning
- Responses API
GLM 5.3 Flash on QuickSilver Pro
GLM 5.3 Flash is Z.ai's fast open-weights sibling of GLM 5.3 — the same 2026 reasoning family distilled to workhorse pricing, with the full 1M-token context window and 131K max output. On QuickSilver Pro it is $0.0284 input / $0.0923 output per million tokens, at parity with the OpenRouter endpoint selected by this route, with no markup or discount. Like GLM 5.3, reasoning is always on for this endpoint and cannot be disabled — control depth with reasoning_effort instead.
- 1M context
- 131K max output
- Tools
- Reasoning · on by default
- Streaming
- No JSON Schema output
You pay · input / output per 1M
$0.0284 / $0.0923
At list price
same rates as OpenRouter
Pricing comparison ($/1M tokens)
Bar length = price relative to the highest price shown
| Provider | Input | Output | vs QSP |
|---|---|---|---|
| QuickSilver Pro | $0.0284 | $0.0923 | — |
| OpenRouter(z-ai/glm-5.3-flash) | $0.0284 | $0.0923 | same |
| OpenAI (GPT-4o mini) | $0.15 | $0.60 | 85% lower |
OpenRouter list source: openrouter.ai/api/v1/models/z-ai/glm-5.3-flash/endpoints
At a glance
- Context
- 1M tokens
- Input / 1M
- $0.0284
- Output / 1M
- $0.0923
- Thinks by default
- Yes
- Max output
- 131K
- Latency · live
- —
- AA Intelligence Index
- 46
High-volume agents and coding at workhorse pricing — the fast open-weights member of Z.ai's newest reasoning family, with a 1M-token context window.
When is GLM 5.3 Flash the right call?
Curated from catalog capabilities and published prices, not a benchmark.
When to use
Reach for GLM 5.3 Flash when you want the GLM 5.3 family's reasoning-and-tools behavior at a fraction of the price: high-volume agent loops, background codegen, summarization over huge working sets in its 1M-token context, and batch pipelines where per-call cost dominates. At $0.0284 / $0.0923 per million tokens it competes with the cheapest tier of the catalog while keeping a 1M-token window — a combination the budget models don't offer. Reasoning is always on; use reasoning_effort to trade depth against latency and cost.
- Long documents and codebases, up to 1M tokens of context
- Agents that call tools
- Multi-step reasoning, on by default
- Responses API clients as well as Chat Completions
When to use something else
For the hardest repo-scale engineering and long-horizon plans, GLM 5.3 is the stronger family pick — same window, deeper model. Reasoning is mandatory here too, so it spends thinking tokens even on trivial turns; if you want a cheap model that replies directly, DeepSeek V4 Flash ($0.06/$0.18) answers without the overhead. If you have prompts tuned on GLM 5.2 or GLM 5.3, A/B before switching down — traces and tool-call cadence differ across the family.
- The task needs a stronger modelGLM 5.3 scores 49 on the AA Intelligence Index (46 here), at 23.835× the output price.GLM 5.3 $0.60 / $2.20 · 1M
- You need strict JSON Schema outputGLM 5.2 lists JSON Schema output, at 23.835× the output price.GLM 5.2 $0.70 / $2.20 · 1M
- You need image inputQwen3.7 Flash accepts image input, at 1.127× the output price.Qwen3.7 Flash $0.024 / $0.104 · 1M
Quickstart (curl)
Get an API key →curl https://api.quicksilverpro.io/v1/chat/completions \
-H "Authorization: Bearer $QSP_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-5.3-flash",
"messages": [{"role": "user", "content": "Hello!"}]
}'import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.quicksilverpro.io/v1",
api_key=os.environ["QSP_KEY"],
)
response = client.chat.completions.create(
model="glm-5.3-flash",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.quicksilverpro.io/v1",
apiKey: process.env.QSP_KEY,
});
const response = await client.chat.completions.create({
model: "glm-5.3-flash",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);curl https://api.quicksilverpro.io/v1/responses \
-H "Authorization: Bearer $QSP_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-5.3-flash",
"input": "Hello!"
}'import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.quicksilverpro.io/v1",
api_key=os.environ["QSP_KEY"],
)
response = client.chat.completions.create(
model="glm-5.3-flash",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)OpenAI-compatible. One-line migration via base_url.
Gateway behavior
Compatibility notes for this model.
prompt- Prompts up to 1M tokens.
max_tokens- Up to 131K output tokens.
- /v1/chat/completions: Supported
- /v1/responses: Supported
- streaming: Supported
- tools: Supported
- image input: Not listed for this model
- JSON Schema: Not listed for this model
- reasoning: Supported
Where the price comes from
- Provider reference
- $0.0284 / $0.0923openrouter.ai/api/v1/models/z-ai/glm-5.3-flash/endpoints
- Catalog verified
- Prompt cache · not compared
- $0.0284 read
usage.cost = prompt_tokens × $0.0284/1M + completion_tokens × $0.0923/1M
FAQ
Common questions
Reasoning is always on — it is mandatory on this endpoint and cannot be disabled (a request that tries to turn it off is rejected upstream). Use reasoning_effort to trade depth against latency and cost. Reasoning tokens bill as output tokens and are drawn from your max_tokens budget before the visible answer, so set max_tokens generously or the reply can come back truncated. If you want a GLM without always-on thinking, use GLM 5.2.
Same family, different weight class: GLM 5.3 Flash is the fast, open-weights variant released 2026-08-26, priced at $0.0284 input / $0.0923 output per million tokens versus $0.60 / $2.20 for GLM 5.3 — about 21.1x lower on input. Both share the 1M-token context window, the 131K max-output limit, and the always-on reasoning contract. Use Flash for volume; step up to GLM 5.3 when task difficulty, not throughput, is the bottleneck.
Yes — GLM 5.3 Flash is an OpenAI-compatible chat completions endpoint on QuickSilver Pro (the Responses API works too). Set base_url=https://api.quicksilverpro.io/v1, paste your QSP key, and use model="glm-5.3-flash". Streaming, tool calling (tool_choice: "auto" — this endpoint rejects forcing a specific function), and usage.cost accounting all work. response_format: json_schema is not enforced on this endpoint yet; if you need strict structured output, use GLM 5.2.
$0.0284 input / $0.0923 output and $0.0284 cache read per million tokens, matching the route-selected OpenRouter endpoint with no markup or discount. The route uses its pinned provider order or the cheapest eligible endpoint under sort: price and its quantization filters; the /models headline can differ. More expensive fallback hosts do not change your price. Use base_url=https://api.quicksilverpro.io/v1 and model="glm-5.3-flash".