Alex Getmanvroege AI-signalen

GLM-5

Z.ai

Uitgebracht Bijgewerkt
Contextvenster
203K
202.752 tokens
Maximale output
131K
131.072 tokens
Invoer / Uitvoer
tekst → tekst

API-prijzen

USD / 1 miljoen tokens

InvoerUitvoerGecachte invoer
  • $1,00
zai-pricing
  • $3,20
zai-pricing
  • $0,20
zai-pricing

Tijdlijn

  1. · Gewichtsversie · bronlink
Prijs invoer / uitvoer · USD / 1 miljoen tokens
ModelBronlinksPrijs invoer / uitvoer
glm-5
standard
zai-pricing
API-route niet beschikbaar
$1,00 / $3,20
  1. · Bronlinks · GitHub search · Z.ai · pulls · smg-project/smg · @hello-alexmcc
    Fragment
    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. · Bronlinks · GitHub search · Z.ai · pulls · milabench/milabench · @bouthilx
    Fragment
    - 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. · Bronlinks · GitHub search · Z.ai · pulls · maximhq/bifrost · @mickgvirtu
    Fragment
    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. · Bronlinks · GitHub search · Z.ai · pulls · anomalyco/models.dev · @models-dev-automation[bot]
  5. · Bronlinks · GitHub search · Z.ai · pulls · ai-dynamo/frontend-crates · @dundysm
  6. · Bronlinks · GitHub search · Z.ai · pulls · ai-dynamo/dynamo · @indrajit96
    Fragment
    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. · Bronlinks · GitHub search · Z.ai · pulls · ROCm/ATOM · @whx-sjtu
  8. · Bronlinks · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @moraxu
    Fragment
    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. · Bronlinks · GitHub search · Z.ai · pulls · NVIDIA/TensorRT-LLM · @kaiyux
    Fragment
    - `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`.
Specificaties

GLM-5 is een AI-model van Z.ai met open gewichten, uitgebracht op 2026-02-12. Het contextvenster is 202.752 tokens. De maximale uitvoer is 131.072 tokens. De prijs per miljoen tokens is $1,00 voor invoer en $3,20 voor uitvoer.

Contextvenster
202.752
Maximale output
131.072
Invoer
tekst
Uitvoer
tekst
Gewichten
Open
Invoer · USD / 1 miljoen tokens
$1,00
Uitvoer · USD / 1 miljoen tokens
$3,20
Gecachte invoer · USD / 1 miljoen tokens
$0,20

Vragen

Wat is het contextvenster van GLM-5?

Het contextvenster is 202.752 tokens. De maximale uitvoer is 131.072 tokens.

Hoeveel kost GLM-5?

De prijs per miljoen tokens is $1,00 voor invoer en $3,20 voor uitvoer. Invoer uit de cache kost $0,20 per miljoen tokens.

Zijn de gewichten van GLM-5 openbaar beschikbaar?

Ja. Z.ai publiceert de modelgewichten.

Wanneer werd GLM-5 uitgebracht?

2026-02-12.