
GLM-5
Z.ai
- কনটেক্সট উইন্ডো
- 203K 2,02,752 টোকেন
- সর্বোচ্চ আউটপুট
- 131K 1,31,072 টোকেন
- ইনপুট / আউটপুট
- লেখা → লেখা
API মূল্য
USD / ১০ লাখ টোকেন
| ইনপুট | আউটপুট | ক্যাশে রাখা ইনপুট |
|---|---|---|
|
|
|
সময়রেখা
- · ওজনের সংস্করণ · প্রমাণের লিঙ্ক
ইনপুট / আউটপুটের দাম · USD / ১০ লাখ টোকেন
| মডেল | প্রমাণের লিঙ্ক | ইনপুট / আউটপুটের দাম |
|---|---|---|
glm-5standard | zai-pricing API রুট অনুপলব্ধ | $1.00 / $3.20 |
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · smg-project/smg · @hello-alexmcc
উদ্ধৃতি
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
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · milabench/milabench · @bouthilx
উদ্ধৃতি
- GLM-5: layout check before launch, loss from the last stage, checkpoint timing and checksums, crash cleanup; `benchmarks/megatron/dev.yaml`, `scripts/trillium/glm/`.
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · maximhq/bifrost · @mickgvirtu
উদ্ধৃতি
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
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · anomalyco/models.dev · @models-dev-automation[bot]
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · ai-dynamo/frontend-crates · @dundysm
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · ai-dynamo/dynamo · @indrajit96
উদ্ধৃতি
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
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · ROCm/ATOM · @whx-sjtu
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @moraxu
উদ্ধৃতি
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`
- · প্রমাণের লিঙ্ক · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @kaiyux
উদ্ধৃতি
- `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`.
প্রযুক্তিগত বিবরণ
GLM-5 হলো Z.ai-এর একটি মডেল এবং এর ওয়েট উন্মুক্ত, প্রকাশিত হয় 2026-02-12। কনটেক্সট উইন্ডো হলো 2,02,752 টোকেন। সর্বোচ্চ আউটপুট হলো 1,31,072 টোকেন। প্রতি ১০ লাখ ইনপুট টোকেনের দাম $1.00 এবং প্রতি ১০ লাখ আউটপুট টোকেনের দাম $3.20।
- কনটেক্সট উইন্ডো
- 2,02,752
- সর্বোচ্চ আউটপুট
- 1,31,072
- ইনপুট
- লেখা
- আউটপুট
- লেখা
- ওয়েট
- উন্মুক্ত
- ইনপুট · USD / ১০ লাখ টোকেন
- $1.00
- আউটপুট · USD / ১০ লাখ টোকেন
- $3.20
- ক্যাশে রাখা ইনপুট · USD / ১০ লাখ টোকেন
- $0.20
প্রশ্ন
GLM-5-এর কনটেক্সট উইন্ডো কত?
কনটেক্সট উইন্ডো হলো 2,02,752 টোকেন। সর্বোচ্চ আউটপুট হলো 1,31,072 টোকেন।
GLM-5-এর দাম কত?
প্রতি ১০ লাখ ইনপুট টোকেনের দাম $1.00 এবং প্রতি ১০ লাখ আউটপুট টোকেনের দাম $3.20। ক্যাশে রাখা ইনপুটের দাম প্রতি ১০ লাখ টোকেনে $0.20।
GLM-5-এর ওয়েট কি সবার জন্য উন্মুক্ত?
হ্যাঁ। Z.ai ওয়েট প্রকাশ করে।
GLM-5 কবে প্রকাশিত হয়?
2026-02-12।