Alex Getmanنشانه‌های زودهنگام هوش مصنوعی

GLM-5

Z.ai

عرضه‌شده به‌روزشده
پنجرهٔ زمینه
203K
202٬752 توکن
حداکثر خروجی
131K
131٬072 توکن
ورودی / خروجی
متن → متن

قیمت‌های API

USD / یک میلیون توکن

ورودیخروجیورودیِ ذخیره‌شده در حافظهٔ نهان
  • $1٫00
zai-pricing
  • $3٫20
zai-pricing
  • $0٫20
zai-pricing

خط زمانی

  1. · نسخهٔ وزن‌ها · پیوند شواهد
قیمت ورودی / خروجی · USD / یک میلیون توکن
مدلپیوندهای شواهدقیمت ورودی / خروجی
glm-5
standard
zai-pricing
مسیر API در دسترس نیست
$1٫00 / $3٫20
  1. · پیوندهای شواهد · GitHub search · Z.ai · pulls · smg-project/smg · @hello-alexmcc
    گزیده
    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
  2. · پیوندهای شواهد · GitHub search · Z.ai · pulls · milabench/milabench · @bouthilx
    گزیده
    - GLM-5: layout check before launch, loss from the last stage, checkpoint timing and checksums, crash cleanup; `benchmarks/megatron/dev.yaml`, `scripts/trillium/glm/`.
  3. · پیوندهای شواهد · GitHub search · Z.ai · pulls · maximhq/bifrost · @mickgvirtu
    گزیده
    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
  4. · پیوندهای شواهد · GitHub search · Z.ai · pulls · anomalyco/models.dev · @models-dev-automation[bot]
  5. · پیوندهای شواهد · GitHub search · Z.ai · pulls · ai-dynamo/frontend-crates · @dundysm
  6. · پیوندهای شواهد · GitHub search · Z.ai · pulls · ai-dynamo/dynamo · @indrajit96
    گزیده
    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
  7. · پیوندهای شواهد · GitHub search · Z.ai · pulls · ROCm/ATOM · @whx-sjtu
  8. · پیوندهای شواهد · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @moraxu
    گزیده
    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`
  9. · پیوندهای شواهد · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @kaiyux
    گزیده
    - `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`.
مشخصات

GLM-5 مدلی از Z.ai است و وزن‌های بازی دارد؛ در 2026-02-12 عرضه شد. پنجرهٔ زمینه 202٬752 توکن است. حداکثر خروجی 131٬072 توکن است. قیمت هر یک میلیون توکن ورودی $1٫00 و هر یک میلیون توکن خروجی $3٫20 است.

پنجرهٔ زمینه
202٬752
حداکثر خروجی
131٬072
ورودی
متن
خروجی
متن
وزن‌ها
باز
ورودی · USD / یک میلیون توکن
$1٫00
خروجی · USD / یک میلیون توکن
$3٫20
ورودیِ ذخیره‌شده در حافظهٔ نهان · USD / یک میلیون توکن
$0٫20

پرسش‌ها

پنجرهٔ زمینهٔ GLM-5 چقدر است؟

پنجرهٔ زمینه 202٬752 توکن است. حداکثر خروجی 131٬072 توکن است.

قیمت GLM-5 چقدر است؟

قیمت هر یک میلیون توکن ورودی $1٫00 و هر یک میلیون توکن خروجی $3٫20 است. قیمت ورودیِ ذخیره‌شده در حافظهٔ نهان به‌ازای هر یک میلیون توکن $0٫20 است.

آیا وزن‌های GLM-5 در دسترس عموم است؟

بله. Z.ai وزن‌ها را منتشر می‌کند.

GLM-5 چه زمانی عرضه شد؟

در 2026-02-12.