Qwen3 8B
Qwenhuggingface.co/Qwen/Qwen3-8B
Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It supports seamless switching between "thinking" mode for math,...
Updated 2 min ago · first seen 11 Sept 2026
model_01M294WWGKJRNVC964ZQYXDZR1
- Parameters
- 8.19B
- T2 · 50 min ago
- Context
- 131.1K tokens
- T2 · 51 min ago
- Released
- 27 Apr 2025
- T2 · 5 min ago
- License
- apache-2.0
- T2 · 5 min ago
- Knowledge cutoff
- Mar 2025
- T2 · 51 min ago
Specification
- Release date
- 27 Apr 2025
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 50 min agomedium
- License
- apache-2.0
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Architecture
- Qwen3ForCausalLM
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Parameters
- 8.19B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 50 min agomedium
- Context window
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Max output
- 8.19K tokens
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Knowledge cutoff
- Mar 2025
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Modalities
- text
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Input modalities
- text
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Tokenizer
- Qwen3
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Base model
- Qwen/Qwen3-8B-Base
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- File size
- 16.4 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Hugging Face repo
- Qwen/Qwen3-8B
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Pipeline tag
- text-generation
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 50 min agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 50 min agomedium
- Downloads
- 12,880,447
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 min agomedium
- Likes
- 1,366
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 min agomedium
- Gated
- No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Hf inference providers
- featherless-ai, nscale
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Instruct type
- qwen3
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Last modified
- 2025-07-26T03:49:13+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 50 min agomedium
- Library name
- transformers
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- License url
- https://huggingface.co/Qwen/Qwen3-8B/blob/main/LICENSE
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Downloads all time
- 120,758,624
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 min agomedium
- Model type
- qwen3
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Openrouter id
- qwen/qwen3-8b
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Supported parameters
- frequency_penalty, include_reasoning, max_tokens, presence_penalty, reasoning, response_format, seed, stop, temperature, tool_choice, tools, top_k, top_p
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Tags
- transformers, safetensors, qwen3, text-generation
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 min agomedium
- Weights dtype
- BF16
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 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
35
Source tiers
T235
Freshest observation
2 min ago
Conflicts
None
Modalities
- Modalities
- text
- Input
- text
- 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
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Max output
- 8.19K tokens
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Knowledge cutoff
- Mar 2025
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
- Tokenizer
- Qwen3
Source:OpenRouter public model & pricing listingT2observed 51 min agomedium
No benchmark results recorded
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| OpenRouterqwen/qwen3-8b | $0.117 | $0.455 | — | — | 131.1K | 51 min ago | openrouter.aiT2 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 1
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- OpenRouter
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 | — | 5.2 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 5.2 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 5.2 GB | Yes |
| Apple M2 Ultra | 4bit | — | 5.2 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 5.2 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 5.2 GB | Yes |
| Apple M1 Ultra | 4bit | — | 5.2 GB | Yes |
| Apple M3 Max | 4bit | — | 5.2 GB | Yes |
| Apple M4 Max | 4bit | — | 5.2 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 5.2 GB | Yes |
| Apple M2 Max | 4bit | — | 5.2 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 5.2 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 5.2 GB | Yes |
| Apple M1 Max | 4bit | — | 5.2 GB | Yes |
| Apple M4 Pro | 4bit | — | 5.2 GB | Yes |
| Apple M3 Pro | 4bit | — | 5.2 GB | Yes |
| Apple M1 Pro | 4bit | — | 5.2 GB | Yes |
| Apple M2 Pro | 4bit | — | 5.2 GB | Yes |
| Apple M4 | 4bit | — | 5.2 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 5.2 GB | Yes |
| Apple M2 | 4bit | — | 5.2 GB | Yes |
| Apple M3 | 4bit | — | 5.2 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 5.2 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 5.2 GB | Yes |
| Apple M1 | 4bit | — | 5.2 GB | Yes |
Ancestors 1
- Qwen3-8B-BaseQwen
Descendants 3
Papers 2
- arXiv:2309.00071Active
- arXiv:2505.09388Active35
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 · Tool calling
Tool callingtool_calling1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| Yes | → 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 8B: release date changed from 2025-04-28 to 2025-04-27
Release date28 Apr 2025→27 Apr 2025huggingfaceOpenRouter lists Qwen3 8B at $0.117 in / $0.455 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/Qwen/Qwen3-8B | model_page | T2· Quality secondary | 2 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/mlx-community/Qwen3-8B-4bit | model_page | T2· Quality secondary | 10 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/amazon/GDN-primed-HQwen3-8B-Instruct | model_page | T2· Quality secondary | 42 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/amazon/GKA-primed-HQwen3-8B-Reasoner | model_page | T2· Quality secondary | 43 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/allenai/Molmo2-8B | model_page | T2· Quality secondary | 45 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=Qwen&p=0&sort=downloads | listing | T2· Quality secondary | 47 min ago | 2 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 51 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.