Updated 2 h ago · first seen 12 Sept 2026
model_01M29XAH078M6SSDF8DVQHM9AD
- Parameters
- 7B
- T2 · 9 h ago
- Released
- 9 Aug 2024
- T2 · 9 h ago
Specification
- Family
- Qwen
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Version
- 2
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Release date
- 9 Aug 2024
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Status
- active
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Openness
- open-weights
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Architecture
- Transformer decoder based on Qwen language model and audio encoder
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Parameters
- 7B
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Mixture of experts
- No
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Max output
- 256 tokens
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Modalities
- audio, text
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Input modalities
- audio, text
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Output modalities
- text
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Languages
- Cantonese, Chinese, English, French, German, Italian, Japanese, Spanish
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Official page
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
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
18
Source tiers
T218
Freshest observation
9 h ago
Conflicts
None
Modalities
- Modalities
- audiotext
- Input
- audiotext
- Output
- text
Capabilities
Tool calling
Unavailable
Structured output
Unavailable
Reasoning
Unavailable
Vision
No
Qwen — official blog · T2
Audio
Yes
Qwen — official blog · T2
Fine-tuning available
Yes
Qwen — official blog · T2
- Max output
- 256 tokens
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
- Languages
- Cantonese, Chinese, English, French, German, Italian, Japanese, Spanish
Source:Qwen — official blogT2observed 9 h agomediumLLM-extracted
No benchmark results recorded
Current prices
No current prices recorded
Price history
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 |
|---|---|---|---|---|
| NVIDIA DGX B200 | 4bit | 1,440 GB | 4.5 GB | Yes |
| Apple M3 Ultra | 4bit | — | 4.5 GB | Yes |
| Mac Studio (Apple M5 Ultra) | 4bit | — | 4.5 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 4.5 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 4.5 GB | Yes |
| Apple M2 Ultra | 4bit | — | 4.5 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 4.5 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 4.5 GB | Yes |
| NVIDIA H200 NVL | 4bit | 141 GB | 4.5 GB | Yes |
| Apple M1 Ultra | 4bit | — | 4.5 GB | Yes |
| Apple M3 Max | 4bit | — | 4.5 GB | Yes |
| Apple M4 Max | 4bit | — | 4.5 GB | Yes |
| Mac Studio (Apple M5 Max) | 4bit | — | 4.5 GB | Yes |
| MacBook Pro (Apple M5 Max) | 4bit | — | 4.5 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 4.5 GB | Yes |
| Apple M2 Max | 4bit | — | 4.5 GB | Yes |
| NVIDIA H100 NVL | 4bit | 94 GB | 4.5 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 4.5 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 4.5 GB | Yes |
| Apple M1 Max | 4bit | — | 4.5 GB | Yes |
| Apple M4 Pro | 4bit | — | 4.5 GB | Yes |
| Mac mini (Apple M5 Pro) | 4bit | — | 4.5 GB | Yes |
| MacBook Pro (Apple M5 Pro) | 4bit | — | 4.5 GB | Yes |
| Apple M3 Pro | 4bit | — | 4.5 GB | Yes |
| Apple M1 Pro | 4bit | — | 4.5 GB | Yes |
| Apple M2 Pro | 4bit | — | 4.5 GB | Yes |
| Apple M4 | 4bit | — | 4.5 GB | Yes |
| iMac (Apple M4) | 4bit | — | 4.5 GB | Yes |
| Mac mini (Apple M6) | 4bit | — | 4.5 GB | Yes |
| MacBook Air (Apple M5) | 4bit | — | 4.5 GB | Yes |
| MacBook Pro (Apple M5) | 4bit | — | 4.5 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 4.5 GB | Yes |
| Apple M2 | 4bit | — | 4.5 GB | Yes |
| Apple M3 | 4bit | — | 4.5 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 4.5 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 4.5 GB | Yes |
| Apple M1 | 4bit | — | 4.5 GB | Yes |
Ancestors 1
This model
Qwen2-Audio
7B params
Quantizations 0
None recorded.
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 · Modalities
Modalitiesmodalities1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| audio, text | → current | current | Qwen — official blogT2 | medium | llm |
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 →
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Qwen — official blog | qwenlm.github.io/blog/qwen2-audio | news | T1· Official | 9 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.