Z.ai · Available now
GLM Flash Latest (2026): the OpenRouter alias for Z.ai’s newest Flash model, what it points to and what it costs
GLM Flash Latest (~z-ai/glm-flash-latest) is an OpenRouter alias, not a Z.ai model: it redirects to the newest model in the GLM Flash family, which OpenRouter’s list shows as GLM 5.3 Flash (320B total, 18B active) on November 19, 2026, billed at the price of whichever host serves each request, so what it runs can change without notice.
Use it now
Chat
≈ 1 credit per reply · See all models
New accounts get 5 free credits once, valid for 7 days. They work on the lower-priced models here; the priciest ones need credits you bought.
Start chatting — 5 free credits- Maker
- Z.ai
- Type
- Text
- Released
- No date given
- OpenRouter id
- ~z-ai/glm-flash-latest
Per million tokens
Price
| - | Input | Output | Source |
|---|---|---|---|
| Z.ai — GLM-5.3-Flash on Z.ai’s own API (today’s target) | $0.15 | $0.50 | docs.z.ai |
| Via OpenRouter (~z-ai/glm-flash-latest) | $0.04 | $0.08 | OpenRouter alias row; the alias has no fixed price of its own, requests are billed at the price of the host that serves them · openrouter.ai |
Checked
Facts, with sources
- OpenRouter’s page for ~z-ai/glm-flash-latest (named "Z.ai: GLM Flash Latest") says the model "always redirects to the latest model in the GLM Flash family". It shows "Latest GLM 5.3 Flash" as the current target, says the alias has no fixed price of its own and that each request is billed at the price of the model that serves it, and gives August 27, 2026 as its release date.Source: openrouter.ai · checked 2026-11-19 · second-hand
- OpenRouter’s documentation says ~author/family-latest slugs always resolve to the newest concrete model in a family and that the response’s model field names the model that actually answered. It also says versions can change at any time, that there is no rollback through the alias, that to pin a version you call the concrete id, and that an alias never resolves to another alias or to a hidden model.Source: openrouter.ai · checked 2026-11-19 · second-hand
- OpenRouter’s public model list (November 19, 2026) gives ~z-ai/glm-flash-latest the alias target z-ai/glm-5.3-flash ("Z.ai: GLM 5.3 Flash") and lists the alias at $0.0399 / $0.0784 per million input / output tokens, with a 1,048,576-token context, 131,072 maximum completion tokens, text, image and video input, and mandatory reasoning (efforts max, high and low).Source: openrouter.ai · checked 2026-11-19 · second-hand
- Z.ai’s GLM-5.3-Flash page calls GLM-5.3-Flash/FlashX the first native multimodal model in the GLM-5 series. GLM-5.3-Flash has 320B total parameters with 18B activated and a hybrid of sparse and linear attention; input is video, image, text and file, output is text, context length 1M and maximum output 128K, and the API model code is glm-5.3-flash.Source: docs.z.ai · checked 2026-11-19
- Z.ai’s release notes list GLM-5.3-Flash under 2026-08-26. The notes do not mention a "latest" alias; the alias is OpenRouter’s.Source: docs.z.ai · checked 2026-11-19
- Z.ai’s pricing page lists GLM-5.3-Flash at $0.15 per million input tokens, $0.03 cached input and $0.50 per million output tokens; GLM-5.3-FlashX at $0.37, $0.075 and $1.25; GLM-5.3 at $1.4, $0.26 and $4.4. It lists no price for a Flash "latest" name.Source: docs.z.ai · checked 2026-11-19
- Z.ai’s page says thinking cannot be disabled on GLM-5.3-Flash (thinking.type supports only enabled) and recommends temperature 1, top_p 0.95 and reasoning_effort max.Source: docs.z.ai · checked 2026-11-19
- Z.ai says GLM-5.3-Flash scores 57 on the Artificial Analysis Intelligence Index v4.1.1 at $0.045 per task (discounted), beats GLM-5.2 on six coding and agentic benchmarks (63.4 vs 46.2 on DeepSWE v1.1, 48.8 vs 26.2 on AutomationBench) while approaching Claude Opus 4.8 overall, and reaches 29.0 against Opus 4.8’s 29.5 on its Z.ai Code Bench v1.0 at max effort. These are Z.ai’s own figures; we have not tested them.Source: docs.z.ai · checked 2026-11-19
- The Hugging Face repository zai-org/GLM-5.3-Flash is public, its model card lists 320B total and 18B active parameters, and the licence field in its metadata reads MIT.Source: huggingface.co · checked 2026-11-19
- The Z.ai pages we read (the GLM-5.3-Flash page and the pricing page) list the API model codes glm-5.3-flash, glm-5.3-flashx and glm-5.3-flashx-legal and do not mention a glm-flash-latest name.Source: docs.z.ai · checked 2026-11-19
What it points to
GLM Flash Latest next to GLM 5.3 Flash, the model it points to today
- On the same list the target z-ai/glm-5.3-flash is priced at $0.15 / $0.50 with 943,717 maximum completion tokens, not the alias’s $0.0399 / $0.0784 and 131,072, although OpenRouter’s documentation says alias rows report the pricing and capabilities of the target they currently resolve to. Both rows list a 1,048,576-token context and text, image and video input.Source: openrouter.ai · checked 2026-11-19 · second-hand
- OpenRouter’s endpoints list for the target (November 19, 2026) shows 42 endpoints from 36 hosts, priced from $0.0399 / $0.0784 (Relace, fp4, 131,072 maximum completion) up to $0.225 / $0.75 (Fireworks), with Z.AI’s own endpoint at $0.15 / $0.50. The alias’s figures match the cheapest endpoint; we do not know why the alias row uses them, and the alias’s own endpoints list is empty.Source: openrouter.ai · checked 2026-11-19 · second-hand
Side by side
The alias, its target on OpenRouter and Z.ai’s own API compared
An alias is a pointer, not a model: OpenRouter keeps one name that always points at the newest model in a family, so code written against it picks up new releases without changes. The table sets the alias next to the model it points to today and to Z.ai’s own API for that model. Most numbers agree, but the price and the maximum completion length differ between the alias row and the target row, and OpenRouter’s documentation says they should match. We report both and do not guess which one a given request will get. If you need the same model next week, use the concrete id; if you want the newest Flash model without editing code, the alias is what OpenRouter offers. Other pages here cover the larger GLM 5.3 and its high-speed variant GLM 5.3 Prime, and Hy4 Flash is another very low-priced multimodal model. On this site a typical reply to GLM Flash Latest costs about 1 credit (a credit is about $0.01 in the $5 pack), billed by its real token cost, at least 1 credit, whenever the chat box above can run the alias.
| What | GLM Flash Latest (alias) | GLM 5.3 Flash on OpenRouter | GLM-5.3-Flash on Z.ai’s API |
|---|---|---|---|
| What it is | OpenRouter alias (id starts with ~), decided by OpenRouter | The Z.ai model z-ai/glm-5.3-flash | Z.ai’s model code glm-5.3-flash |
| Price per million tokens, input / output | $0.0399 / $0.0784 (list row; billed at the serving host’s price) | $0.15 / $0.50 (list row) | $0.15 / $0.50 (cached input $0.03) |
| Context window | 1,048,576 tokens | 1,048,576 tokens | 1M tokens |
| Maximum output | 131,072 tokens | 943,717 tokens | 128K tokens |
| Input | Text, image and video | Text, image and video | Video, image, text and file |
| Reasoning | Mandatory; efforts max, high, low | Mandatory; efforts max, high, low | Thinking cannot be disabled |
| Hosts | None of its own; requests go to the target’s hosts | 42 endpoints from 36 hosts, $0.0399 / $0.0784 to $0.225 / $0.75 | Z.ai’s own endpoint (Z.AI, fp8) |
| Listed or released | Listed August 27, 2026 | Listed August 26, 2026 | Released August 26, 2026 |
| Typical reply on this site (our arithmetic) | about 1 credit, when the chat can run the alias | about 1 credit at the list price (not a separate page here) | Not run here: Z.ai’s own API is not used by the chat |
Source: openrouter.aiSource: openrouter.aiSource: docs.z.aiSource: docs.z.aiSource: openrouter.aiSource: openrouter.aiSource: docs.z.ai
The Flash line
The GLM Flash models on Z.ai’s price page, and what the two aliases point to
- Z.ai’s page describes GLM-5.3-FlashX as the version with 200 tokens/s inference speed and GLM-5.3-FlashX-Legal as a model tuned for legal workflows such as legal research, contract review and drafting; it says GLM-5.3-Flash is available on the GLM Coding Plan with 3× the quota of GLM-5.3, and FlashX is not yet on the plan.Source: docs.z.ai · checked 2026-11-19
- OpenRouter’s model list (November 19, 2026) gives the sister alias ~z-ai/glm-latest the target z-ai/glm-5.3, with the same $0.0699 / $1.4 price, 1,048,576-token context and 943,718 maximum completion tokens as z-ai/glm-5.3 on the same list.Source: openrouter.ai · checked 2026-11-19 · second-hand
Z.ai prices
GLM-5.3-Flash, FlashX, FlashX-Legal and GLM-5.3 compared
The word "Flash" covers several Z.ai models at different prices, so it matters which one a name points to. Z.ai prices GLM-5.3-Flash at $0.15 / $0.50 per million tokens, the faster FlashX at $0.37 / $1.25 and the full GLM-5.3 at $1.4 / $4.4. OpenRouter keeps a second alias, ~z-ai/glm-latest, and its list points that one at GLM 5.3, not at a Flash model, so the two "latest" names lead to different models with very different output prices. The existing page for GLM 5.3 covers the full model.
| What | GLM-5.3-Flash | GLM-5.3-FlashX | GLM-5.3-FlashX-Legal | GLM-5.3 |
|---|---|---|---|---|
| Z.ai price per million tokens, input / output | $0.15 / $0.50 | $0.37 / $1.25 | $0.37 / $1.25 | $1.4 / $4.4 |
| Cached input price | $0.03 | $0.075 | $0.075 | $0.26 |
| What Z.ai says it is | First native multimodal GLM-5 model; 320B total, 18B active | The faster version, 200 tokens/s | Tuned for legal workflows | Z.ai’s latest flagship model |
| Does a "latest" alias point to it today? | Yes: ~z-ai/glm-flash-latest | No alias points to it | No alias points to it | Yes: ~z-ai/glm-latest |
Source: docs.z.aiSource: docs.z.aiSource: docs.z.aiSource: openrouter.ai
Nearest available
Models you can use now instead
GLM 5.3
Z.ai$0.08 / $3.6 (per million tokens)
- A secondary report (DataNorth) says Z.ai released GLM-5.3 on August 14, 2026; Z.ai’s docs page calls it "Z.ai’s latest flagship model" but states no date.Source: datanorth.ai · checked 2026-11-15 · second-hand
- Z.ai’s docs say GLM-5.3 advances complex software engineering and agent capabilities, with an emphasis on coding and cybersecurity performance.Source: docs.z.ai · checked 2026-11-15
- Context window 1M tokens, maximum output 128K tokens; text-only input.Source: docs.z.ai · checked 2026-11-15
Hy4 Flash
Tencent$0.04 / $0.14 (per million tokens)
- Tencent’s Hy4 flash announcement, dated November 16, 2026, presents it as the newest member of the Hy model family. Tencent recommends it for most everyday tasks (deep research, document processing, PowerPoint generation, simple front-end work, computer use) and points complex tasks such as long-horizon software engineering to Hy4. Tencent Cloud’s API guide calls it "a leaner, more efficient everyday workhorse".Source: tencent.com · checked 2026-11-17
- The Tencent announcement is dated November 16, 2026, and Tencent Cloud’s model, pricing and API pages for hy4-flash were last updated the same day. The announcement says only that Hy4 flash is available now; it gives no separate release date or version name.Source: tencent.com · checked 2026-11-17
- Tencent says Hy4 flash has 224B total parameters (including 64B Engram parameters), 12B active parameters, a context of up to 1M tokens and multimodal input; the announcement does not state a maximum output or the output types. It says the KV cache is 1/6 the size of Hy4’s and that decoding exceeds 150 tokens per second. These are the maker’s own figures, not tested by us.Source: tencent.com · checked 2026-11-17
FAQ
Questions
What is GLM Flash Latest?
GLM Flash Latest, id ~z-ai/glm-flash-latest, is an alias that OpenRouter created; it is not a model Z.ai released. OpenRouter says it always redirects to the latest model in the GLM Flash family, so a request sent to it is forwarded to whichever Flash model OpenRouter currently treats as newest. The response reports the concrete model that answered.
Which model does GLM Flash Latest point to today?
According to OpenRouter’s model list on November 19, 2026, GLM 5.3 Flash (z-ai/glm-5.3-flash); the alias’s own page also shows GLM 5.3 Flash as the latest model. Z.ai’s release notes date GLM-5.3-Flash 2026-08-26. The faster FlashX and the legal FlashX-Legal models appear on Z.ai’s price page, but OpenRouter’s list does not point the alias at them.
Why is the alias price different from the GLM 5.3 Flash price?
OpenRouter lists the alias at $0.0399 / $0.0784 per million tokens and the target at $0.15 / $0.50, although its documentation says alias rows report the target’s pricing. Today’s endpoint list for the target runs from $0.0399 / $0.0784 (Relace) to $0.225 / $0.75 (Fireworks), and the alias matches the cheapest. We do not know why the two rows use different prices, so treat neither as a promise of what one request costs.
How much does GLM Flash Latest cost?
It has no price of its own: each request is billed at the price of the host that serves it, which today lies between $0.0399 / $0.0784 and $0.225 / $0.75 per million input / output tokens, while Z.ai’s own API charges $0.15 / $0.50 (cached input $0.03). On this site, whenever the chat can run the alias, a typical reply costs about 1 credit (a credit is about $0.01 in the $5 pack), billed by its real token cost, at least 1 credit; the box on this page says if it cannot.
Can the model behind GLM Flash Latest change without notice?
Yes. OpenRouter’s documentation says versions can change at any time, that the alias always goes to the newest eligible model and that there is no rollback through the alias. If a newer Flash model appears, requests move to it, so a test you ran today may not hold tomorrow. To pin a version call the concrete id z-ai/glm-5.3-flash, or read the model field of each response.
How is it different from GLM Latest and GLM 5.3?
OpenRouter’s list points ~z-ai/glm-latest at z-ai/glm-5.3, the full flagship, which has its own page here: GLM 5.3. GLM Flash Latest points at the Flash line instead, which Z.ai prices at $0.15 / $0.50 against $1.4 / $4.4. GLM 5.3 Prime is a separate high-speed variant of GLM 5.3 on OpenRouter, not part of the Flash line.
What is GLM 5.3 Flash good at?
Z.ai describes it as a native multimodal model for coding, office documents and financial research. By its own figures it scores 63.4 on DeepSWE v1.1 against GLM-5.2’s 46.2 and 29.0 against Claude Opus 4.8’s 29.5 on Z.ai Code Bench at max effort. These are Z.ai’s numbers; we have not tested them.
Is this the official Z.ai website?
No. This page is unofficial and not affiliated with or endorsed by Z.ai. Facts link to the pages we read.