
GLM-5
Z.ai
- コンテキスト長
- 203K 202,752 トークン
- 最大出力
- 131K 131,072 トークン
- 入力 / 出力
- テキスト → テキスト
API料金
USD / 100万トークン
| 入力 | 出力 | キャッシュ入力 |
|---|---|---|
|
|
|
時系列
- · 重みのバージョン · 参照リンク
入力/出力料金 · USD / 100万トークン
| モデル | 参照リンク | 入力/出力料金 |
|---|---|---|
glm-5standard | zai-pricing APIルートは利用できません | $1.00 / $3.20 |
- · 参照リンク · GitHub search · Z.ai · pulls · smg-project/smg · @hello-alexmcc
抜粋
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
- · 参照リンク · GitHub search · Z.ai · pulls · milabench/milabench · @bouthilx
抜粋
- GLM-5: layout check before launch, loss from the last stage, checkpoint timing and checksums, crash cleanup; `benchmarks/megatron/dev.yaml`, `scripts/trillium/glm/`.
- · 参照リンク · GitHub search · Z.ai · pulls · maximhq/bifrost · @mickgvirtu
抜粋
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
- · 参照リンク · GitHub search · Z.ai · pulls · anomalyco/models.dev · @models-dev-automation[bot]
- · 参照リンク · GitHub search · Z.ai · pulls · ai-dynamo/frontend-crates · @dundysm
- · 参照リンク · GitHub search · Z.ai · pulls · ai-dynamo/dynamo · @indrajit96
抜粋
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
- · 参照リンク · GitHub search · Z.ai · pulls · ROCm/ATOM · @whx-sjtu
- · 参照リンク · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @moraxu
抜粋
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`
- · 参照リンク · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @kaiyux
抜粋
- `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`.
仕様
GLM-5 は Z.ai のAIモデルです(重み公開)。リリース日:2026-02-12。 コンテキストウィンドウは202,752 トークンです。 最大出力は131,072 トークンです。 100万トークンあたりの料金は、入力が$1.00、出力が$3.20です。
- コンテキスト長
- 202,752
- 最大出力
- 131,072
- 入力
- テキスト
- 出力
- テキスト
- 重み
- 公開
- 入力 · USD / 100万トークン
- $1.00
- 出力 · USD / 100万トークン
- $3.20
- キャッシュ入力 · USD / 100万トークン
- $0.20
よくある質問
GLM-5 のコンテキスト長は?
コンテキストウィンドウは202,752 トークンです。 最大出力は131,072 トークンです。
GLM-5 の料金は?
100万トークンあたりの料金は、入力が$1.00、出力が$3.20です。 キャッシュ入力は100万トークンあたり $0.20。
GLM-5 の重みは公開されていますか?
はい。Z.ai が重みを公開しています。
GLM-5 のリリース日は?
2026-02-12。