GLM-5
Z.ai
- Context window
- 203K 202,752 tokens
- Max output
- 131K 131,072 tokens
- Input / Output
- text → text
API pricing
USD / 1M tokens
| Input | Output | Cached input |
|---|---|---|
|
|
|
Timeline
- · Weights version · evidence link
Input / output price · USD / 1M tokens
| Model | Evidence links | Input / output price |
|---|---|---|
glm-5standard | zai-pricing API route unavailable | $1.00 / $3.20 |
- · Evidence links · GitHub search · Z.ai · pulls · smg-project/smg · @hello-alexmcc
Excerpt
Atria-Dawn, `glm`, beside Atria-Dawn-Preview: the template writes the GLM-5 shape (`[gMASK]<sop>`, a reasoning-effort system turn, `<|system|>` / `<|user|>` / `<|assistant|>` / `<|observation|>` turns), the tools
- · Evidence links · GitHub search · Z.ai · pulls · milabench/milabench · @bouthilx
Excerpt
- GLM-5: layout check before launch, loss from the last stage, checkpoint timing and checksums, crash cleanup; `benchmarks/megatron/dev.yaml`, `scripts/trillium/glm/`.
- · Evidence links · GitHub search · Z.ai · pulls · maximhq/bifrost · @mickgvirtu
Excerpt
is not `schemas.Anthropic`, and the served model -- `glm-5`, `Kimi-K2`, ... -- is neither opus-4.8 nor Fable), so Claude Code's per-turn reminders were hoisted every
- · Evidence links · GitHub search · Z.ai · pulls · anomalyco/models.dev · @models-dev-automation[bot]
- · Evidence links · GitHub search · Z.ai · pulls · ai-dynamo/frontend-crates · @dundysm
- · Evidence links · GitHub search · Z.ai · pulls · ai-dynamo/dynamo · @indrajit96
Excerpt
for any DSA model (including DeepSeek V3.2 and GLM-5), or matching `flashmla_kv`/`trtllm` kernels on K3.1's sharded DCP path. Kernels SGLang has not resolved at CLI
- · Evidence links · GitHub search · Z.ai · pulls · ROCm/ATOM · @whx-sjtu
- · Evidence links · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @moraxu
Excerpt
class is registered for both `DeepseekV3ForCausalLM` and `DeepseekV32ForCausalLM`; GLM-5 (`GlmMoeDsaForCausalLM`) and Mistral Large 3 subclass it, so exact-class matching excludes them. Kimi K2 declares `DeepseekV3ForCausalLM`
- · Evidence links · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @kaiyux
Excerpt
- `modeling_glm5_next_vision.py`: the GLM-5 input processor takes the prefix tokenization cache (the fourth commit). Before, it tokenized every text prompt in full and ignored `enable_tokenization_cache`.
Specifications
GLM-5 is a model from Z.ai with open weights, released on 12 Feb 2026. The context window is 202,752 tokens. The maximum output is 131,072 tokens. Pricing is $1.00 per million input tokens and $3.20 per million output tokens.
- Context window
- 202,752
- Max output
- 131,072
- Input
- text
- Output
- text
- Weights
- Open
- Input · USD / 1M tokens
- $1.00
- Output · USD / 1M tokens
- $3.20
- Cached input · USD / 1M tokens
- $0.20
Questions
What is the context window of GLM-5?
The context window is 202,752 tokens. The maximum output is 131,072 tokens.
How much does GLM-5 cost?
Pricing is $1.00 per million input tokens and $3.20 per million output tokens. Cached input is $0.20 per million.
Are the weights of GLM-5 publicly available?
Yes. Z.ai publishes the weights.
When was GLM-5 released?
On 12 Feb 2026.