Qwen3.8 27B
Qwenhuggingface.co/Qwen/Qwen3.8-27B
Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...
Updated 10 min ago · first seen 11 Sept 2026
model_01M294WVR4JWTTVTSEYX2E0VCE
- Parameters
- 27.8B
- T2 · 58 min ago
- Context
- 1M tokens
- T2 · 1 h ago
- Released
- 5 Aug 2026
- T2 · 10 min ago
- License
- apache-2.0
- T2 · 10 min ago
Specification
- Release date
- 5 Aug 2026
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
- License
- apache-2.0
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Architecture
- Qwen3_5ForConditionalGeneration
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Parameters
- 27.8B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
- Context window
- 1M tokens
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Max output
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Modalities
- image, text, video
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Input modalities
- image, text, video
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Tokenizer
- Qwen
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- File size
- 55.6 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Hugging Face repo
- Qwen/Qwen3.8-27B
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Pipeline tag
- image-text-to-text
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
- Downloads
- 7,563,763
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Likes
- 14,752
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Gated
- No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Groq model id
- qwen/qwen3.8-27b
Source:Groq — docs & pricingT1observed 1 h agohigh
- Groq status
- preview
Source:Groq — docs & pricingT1observed 1 h agohigh
- Hf inference providers
- cerebras, deepinfra, featherless-ai, novita, ovhcloud
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Last modified
- 2026-08-14T15:00:01+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
- Library name
- transformers
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Downloads all time
- 7,563,763
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Model type
- qwen3_5
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Openrouter id
- qwen/qwen3.8-27b
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Supported parameters
- frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_a, top_k, top_logprobs, top_p
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Tags
- transformers, safetensors, qwen3_5, image-text-to-text
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
- Weights dtype
- BF16
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 10 min agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
34
Source tiers
T1T22 / 32
Freshest observation
10 min ago
Conflicts
None
Modalities
- Modalities
- imagetextvideo
- Input
- imagetextvideo
- Output
- text
Capabilities
Tool calling
Yes
OpenRouter public model & pricing listing · T2
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Yes
OpenRouter public model & pricing listing · T2
Vision
Yes
OpenRouter public model & pricing listing · T2
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 1M tokens
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Max output
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
- Tokenizer
- Qwen
Source:OpenRouter public model & pricing listingT2observed 1 h agomedium
No benchmark results recorded
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| OpenRouterqwen/qwen3.8-27b | $0.42 | $3 | $0.085 | — | 1M | 1 h ago | openrouter.aiT2 |
| GroqCloudqwen/qwen3.8-27b | $0.80 | $4 | — | — | 131K | 1 h ago | console.groq.comT1 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 2
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- OpenRouter
- GroqCloud
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| Apple M3 Ultra | 4bit | — | 16.5 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 16.5 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 16.5 GB | Yes |
| Apple M2 Ultra | 4bit | — | 16.5 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 16.5 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 16.5 GB | Yes |
| Apple M1 Ultra | 4bit | — | 16.5 GB | Yes |
| Apple M3 Max | 4bit | — | 16.5 GB | Yes |
| Apple M4 Max | 4bit | — | 16.5 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 16.5 GB | Yes |
| Apple M2 Max | 4bit | — | 16.5 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 16.5 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 16.5 GB | Yes |
| Apple M1 Max | 4bit | — | 16.5 GB | Yes |
| Apple M4 Pro | 4bit | — | 16.5 GB | Yes |
| Apple M3 Pro | 4bit | — | 16.5 GB | Yes |
| Apple M1 Pro | 4bit | — | 16.5 GB | Yes |
| Apple M2 Pro | 4bit | — | 16.5 GB | Yes |
| Apple M4 | 4bit | — | 16.5 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 16.5 GB | Yes |
| Apple M2 | 4bit | — | 16.5 GB | Yes |
| Apple M3 | 4bit | — | 16.5 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 16.5 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 16.5 GB | Yes |
| Apple M1 | 4bit | — | 16.5 GB | No |
Ancestors 0
None recorded.
This model
Qwen3.8 27B
27.8B params
Quantizations 10
- amd/Qwen3.8-27B-Quark-AWQ-INT4-W4A166.47B
- amd/Qwen3.8-27B-Quark-AWQ-MXFP415.6B
- mlx-community/Qwen3.8-27B-4bit27.4B
- mlx-community/Qwen3.8-27B-8bit27.4B
- perplexity-ai/pplx-computer-qwen-3-8-27b-dflash2-gguf-2026082627B
- pplx-computer-qwen-3-8-27b-dflash2-2026082418.8B
- pplx-qwen-3-8-27b-dflash2-2026081918.8B
- Qwen/Qwen3.8-27B-FP827.8B
- unsloth/Qwen3.8-27B-GGUF27B
- unsloth/Qwen3.8-27B-NVFP419.9B
Descendants 0
None recorded.
Papers 0
No papers linked yet.
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Supported parameters
Supported parameterssupported_parameters1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_a, top_k, top_logprobs, top_p | → current | current | OpenRouter public model & pricing listingT2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
Qwen3.8 27B: release date changed from 2026-08-14 to 2026-08-05
Release date14 Aug 2026→5 Aug 2026huggingfaceGroqCloud lists Qwen3.8 27B at $0.8 in / $4 out per 1M tokens
groq_pricingOpenRouter lists Qwen3.8 27B at $0.42 in / $3 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Groq — docs & pricing | console.groq.com/docs/models | model_docs | T1· Official | 57 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/unsloth/Qwen3.8-27B-NVFP4 | model_page | T2· Quality secondary | 2 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/unsloth/Qwen3.8-27B-GGUF | model_page | T2· Quality secondary | 2 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/Qwen/Qwen3.8-27B | model_page | T2· Quality secondary | 10 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/Qwen/Qwen3.8-27B-FP8 | model_page | T2· Quality secondary | 10 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/perplexity-ai/pplx-qwen-3-8-27b-dflash2-20260819 | model_page | T2· Quality secondary | 11 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/perplexity-ai/pplx-computer-qwen-3-8-27b-dflash2-gguf-20260826 | model_page | T2· Quality secondary | 11 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/perplexity-ai/pplx-computer-qwen-3-8-27b-dflash2-20260824 | model_page | T2· Quality secondary | 11 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/mlx-community/Qwen3.8-27B-8bit | model_page | T2· Quality secondary | 19 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/mlx-community/Qwen3.8-27B-4bit | model_page | T2· Quality secondary | 19 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/amd/Qwen3.8-27B-Quark-AWQ-MXFP4 | model_page | T2· Quality secondary | 50 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/amd/Qwen3.8-27B-Quark-AWQ-INT4-W4A16 | model_page | T2· Quality secondary | 50 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=Qwen&p=0&sort=downloads | listing | T2· Quality secondary | 55 min ago | 2 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 1 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.