Qianfan-VL-70B
- Reasoning
- Open weights
Prices
1 offering · prices as listedQianfan-VL-70B| Conditions/ 1K tokens | Input | Output |
|---|---|---|
| Standard past price, from the vendor’s page archived Jun 22, 2026 | ¥0.008 | ¥0.024 |
| Batch past price, from the vendor’s page archived Jun 22, 2026 | ¥0.0032 | ¥0.0096 |
Prices from 百度智能云千帆：模型价格 (opens in a new tab) (checked Oct 5, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Family
- qianfan-vl
- Model ID
Not stated on Baidu’s pages: max input, output modalities, reasoning efforts, reasoning can be off, max thinking budget, tool use, structured output, built-in tools, knowledge cutoff, licence, aliases.
Run locally
Weights: baidu/Qianfan-VL-70B- Parameters 71.8B
- Dense
- Native precision BF16
- Layers 80
- KV cache per token 320 KB
- Config positions 32K (config.json max_position_embeddings)
- Needs
- ≈ 39.8 GB (weights 35.5 + KV cache 2.5 + overhead 1.8)
- Usable
- 21.6 GB (90% of its memory)
- In system memory
- 18.7 GB · 38/80 layers on the GPU (--n-gpu-layers 38)
- Generation speed
- 2–3 tok/s
- First token
- about 2.32–3.48 s for a 2K prompt
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| BF16 (native)maker’s file | 133.8 GB | ≈ 143.0 | ✕Doesn’t fit | — |
| Q8_0 | 71.1 GB | ≈ 77.1 | ✕Doesn’t fit | — |
| FP8 | 66.9 GB | ≈ 72.8 | ✕Doesn’t fit | — |
| Q6_K | 54.9 GB | ≈ 60.2 | ✕Doesn’t fit | — |
| Q5_K_M | 47.6 GB | ≈ 52.5 | ✕Doesn’t fit | — |
| Q5_0 | 46.0 GB | ≈ 50.8 | ✕Doesn’t fit | — |
| Q4_K_M | 40.6 GB | ≈ 45.2 | ✕Doesn’t fit | — |
| Q4_0 | 37.6 GB | ≈ 42.0 | ○Needs system memory 2–3 tok/s | — |
| IQ4_XS | 35.5 GB | ≈ 39.8 | ○Needs system memory 2–3 tok/s | — |
| Q3_K_Mquality drops | 32.7 GB | ≈ 36.8 | ○Needs system memory 2–4 tok/s | — |
| IQ3_XXSquality drops | 25.6 GB | ≈ 29.4 | ○Needs system memory 4–7 tok/s | — |
| Q2_Kquality drops | 25.1 GB | ≈ 28.8 | ○Needs system memory 5–7 tok/s | — |
| IQ2_XXSquality drops | 17.2 GB | ≈ 20.6 | ●Runs well 24–36 tok/s | 11K |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): GPU NVIDIA H100 SXM; unified memory Apple M5 Max (40-core GPU) · 64 GB.
Run commands
The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by usThis model isn’t on any leaderboard we cite yet.
Changes
All changes →No changes recorded yet.
Sources
- 百度智能云千帆：模型价格in Chinese (opens in a new tab) cloud.baidu.com · checked Oct 5, 2026
- Qianfan-VL系列模型重磅开源！全尺寸领域增强效果优异，全自研芯片计算！ | 百度智能云in Chinese (opens in a new tab) cloud.baidu.com · checked Nov 5, 2026
- baidu/Qianfan-VL-70B model card (opens in a new tab) huggingface.co · checked Nov 3, 2026
- baidu/Qianfan-VL-70B config.json (opens in a new tab) huggingface.co · checked Nov 3, 2026
- baidu/Qianfan-VL-70B files (opens in a new tab) huggingface.co · checked Nov 4, 2026
- 模型更新记录in Chinese (opens in a new tab) ai.baidu.com · checked Nov 5, 2026
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
Baidu · China · cloud.baidu.com
Related models
Snapshots, the same family and other models from this provider