Qwen3.8
Alibaba
Garis masa
- · Pautan bukti · GitHub search · Alibaba · commits · onprem-ai/deployment-catalog · @tomas-polach
- · Pautan bukti · GitHub search · Alibaba · commits · finereli/pyramid-mcp · @finereli
- · Pautan bukti · GitHub search · Alibaba · commits · fdietze/pi-infinite-context · @fdietze
- · Siaran Alex · keluarga model · GitHub search · Alibaba · commits · agentgateway/agentgateway · @github-actions[bot]
Petikan
json.providers.nebius.models["Qwen/Qwen3.8-27B"].name = "Qwen3.8 27B";
- · Pautan bukti · GitHub search · Alibaba · commits · aabhay007/Personal-AI-Core · @aabhay007
- · Pautan bukti · GitHub search · Alibaba · commits · QuantumKuba/Diffactly · @QuantumKuba
- · Pautan bukti · GitHub search · Alibaba · commits · Hallo-ich-bin-PeppaPig/Singularity · @Hallo-ich-bin-PeppaPig
- · Pautan bukti · GitHub search · Alibaba · commits · Azure/azureml-assets · @hanbitmyths
Petikan
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models (#5452)
- · Siaran Alex · keluarga model · GitHub search · Alibaba · pulls · xadupre/devdash-agents · @xadupre
Petikan
- Real Qwen3.8 27B on H200-101, physical CUDA 7: streaming generation and a follow-up turn succeeded. Used a temporary configuration with 16 KV blocks and
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] qwen38_flash package entry: register() and READMEs
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] MTP draft model
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] KV offload pool regions
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Runner input adapter
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Hyper-connection layers
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] GDN layer input preparation
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Pooled KV cache materialization
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] PD transfer: identity, manifest and Raiden connector
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Config hook, geometry checks and pooled cache layout
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] PLE feature: tables, host lookup, storage, offload and pooled PLE
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Sequence-parallel prefill backbone
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] QSA attention layer
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Model-owned GDN op and layer (no register_oot)
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Per-group Mamba index accessors
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Mamba tail fold: export prefill checkpoints from inside the step
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] In-place rows update op
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] PLE Torch-to-JAX ops
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] QSA Torch-to-JAX ops
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Torch bridge for the QSA kernels
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] GDN input preparation kernel
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Unified KV pool kernels
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] PLE lookup and gate kernels
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Hyper-connection kernels
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Pooled QSA kernel over the unified KV pool
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] QSA KV write kernels (fused, main, compressed)
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] QSA union-walk kernels (plan, dense, SparseCore gather, prefill)
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] Qwen3.8-Flash model and registration hooks
- · Pautan bukti · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Petikan
[Model][Qwen3.8] QSA main-KV decode context parallelism over rank pairs (DCP-2) kernels
- · Pautan bukti · GitHub search · Alibaba · pulls · vjmichael/Lu-AI-GUI · @vjmichael
Petikan
- Physical interactive owner Windows/Radeon Qwen3.8, Vulkan VRAM/TFFB TTFT p50/p95, and actual local installed MMProj; hosted CPU small models do not substitute.
- · Pautan bukti · GitHub search · Alibaba · pulls · vitessio/vitess · @jdoupe
- · Pautan bukti · GitHub search · Alibaba · pulls · nearai/cvm-compose-files · @PierreLeGuen
- · Pautan bukti · GitHub search · Alibaba · pulls · michsindlinger/specwright · @michsindlinger
- · Siaran Alex · keluarga model · GitHub search · Alibaba · pulls · SceneWorks/SceneWorks · @michaeltrefry
Petikan
SW-7119: Qwen3.8 27B describer model and tier in core, worker, API and catalog
- · Siaran Alex · keluarga model · GitHub search · Alibaba · pulls · NVIDIA/hyperflare-inference · @Raydius
Petikan
The pinned Qwen3.8 Flash Next checkpoint trains a 262,144-token context (theta 10M, interleaved MRoPE 11/11/10). This adds a declared serving variant of the same weights
- · Pautan bukti · GitHub search · Alibaba · pulls · FlashML-org/FreeToken · @zihaomu
- · Siaran Alex · keluarga model · GitHub search · Alibaba · pulls · BuQQzz/SHIELD2.0 · @BuQQzz
Petikan
Makes Qwen3.8 27B (IQ1_M) a working coding model on an 8 GB card (RTX 2070 SUPER) through SHIELD's own llama-server.
- · Pautan bukti · GitHub search · Alibaba · pulls · Azure/azureml-assets · @hanbitmyths
Petikan
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models
- · Pautan bukti · GitHub search · Alibaba · pulls · 0xburden/modelship · @0xburden
Petikan
- Reviewed immutable catalog of 25 standalone Qwen3.8 GGUF quantizations; wizard displays sizes, numeric/letter/full-value selection and a visible UD-Q4_K_M default. No hardware-driven quant substitution; flags
- · Pautan bukti · GitHub · QwenLM/qwen-code · PR
Petikan
- The Standard preset has the same qwen3.8 effort and GLM thinking gaps and will be aligned in a follow-up
- · Siaran Alex · keluarga model · Alex Getman
Petikan
🚨 Qwen3.8 Omni Flash is out It can see, hear and understand video, then actually act on it Give it hours of footage and it can find the right moments, translate, edit or turn it into notes. Give it a meeting and it can pull out tasks and start working on them 1M context, API from $0.43
- · Siaran Alex · keluarga model · Alex Getman
Petikan
🚨 QWEN3.8 FLASH NEXT IS OUT The leaked specs were real: • 125B multimodal MoE • Only 6B active • 262K native context, up to 1M Qwen says it beats Claude Opus 4.6 across coding, agentic tasks, tool use and several multimodal benchmarks.
- · Siaran Alex · keluarga model · Alex Getman
Petikan
🚨 QWEN3.8 FLASH NEXT DROPS TODAY A ModelScope description that was quickly taken down reportedly revealed: • 125B main parameters • 51B N gram embeddings • Only 6B active per token • Multimodal MoE architecture The model is also an official preview of the upcoming Qwen4 architecture.
- · Siaran Alex · Alex Getman
Petikan
⚡️ Qwen 3.8 27B beat GPT 5.6 Terra This happened in the AA Agentic Index, a benchmark that tests how well models handle agentic tasks. And you can run Qwen 3.8 27B locally on a single RTX 3090.
- · Siaran Alex · Alex Getman
Petikan
⚡️ Alibaba released Qwen3.8’s weights A model at Claude Opus 4.6 level, a flagship just six months ago, can now run on a single RTX 4090. Qwen3.8 27B uses around 17 GB of VRAM when quantized and should deliver ~50 tok/s.
- · Siaran Alex · Alex Getman
Petikan
Alibaba just teased Qwen3.8 with 2.4T parameters This is the new Qwen flagship, with open weights coming soon. Alibaba ranks it right below Claude Fable 5. Public benchmarks are still missing, so nobody really knows how close it actually is to Fable, or how it stacks up against GPT-5.6 Sol and Kimi K3. Pricing is the fun part. Kimi K3 costs $3/$15, so Qwen3.8 will probably land in the same range. The preview is already live in Qoder. It uses 10x fewer credits during the day and 50x fewer at night, so you can basically test it for free.
Spesifikasi
Radar · Qwen3.8 · Tarikh keluaran tidak diketahui
Soalan
Bilakah Qwen3.8 dikeluarkan?
Tarikh keluaran tidak diketahui