---
source: 'https://howaiworks.ai/models/ernie'
section: models
title: ERNIE 5.1
description: >-
  Baidu's closed-weight flagship, released May 9, 2026. A sparse
  Mixture-of-Experts model with a 128K context — the first ERNIE whose weights
  stay unpublished.
tags:
  - ERNIE
  - Wenxin
  - Baidu
  - Language Model
  - Large Language Model
  - MoE
  - Closed Weights
  - Proprietary
  - Chinese AI
  - Latest
category: Language Models
developer: Baidu
developerWebsite: 'https://home.baidu.com/'
modelType: Language Model
releaseDate: '2026-05-09'
lastUpdated: '2026-09-06'
license: Proprietary
contextWindow: 128K tokens
officialWebsite: 'https://ernie.baidu.com/'
docsPage: 'https://cloud.baidu.com/doc/qianfan/s/rmh4stp0j'
pricingPage: 'https://cloud.baidu.com/doc/qianfan/s/wmh4sv6ya'
---

# ERNIE 5.1

> Baidu's closed-weight flagship, released May 9, 2026. A sparse Mixture-of-Experts model with a 128K context — the first ERNIE whose weights stay unpublished.

## Overview

ERNIE 5.1 (文心 5.1) is Baidu's closed-weight flagship [large language model](https://howaiworks.ai/glossary/large-language-model), **released on May 9, 2026**. Baidu's release post opens: "Today, we officially release ERNIE 5.1. While inheriting the pre-training foundation of ERNIE 5.0, it compresses total parameters to approximately one-third and active parameters to approximately one-half, **achieving leading foundational performance at its model scale using only about 6% of the pre-training cost of comparable models.**"

The headline is efficiency, not scale. ERNIE 5.0, announced as a "2.4 Trillion-Parameter Unified Multimodal Foundation Model" with a "<3% Activation Rate," was the ceiling; ERNIE 5.1 is a smaller [mixture-of-experts](https://howaiworks.ai/glossary/mixture-of-experts) model carved out of the same pre-training run. Baidu calls the technique **Multi-Dimensional Elastic Pre-Training**, combining elastic depth, elastic width / expert capacity, and elastic sparsity — the last defined as: "Through a variable Top-k routing mechanism, the number of activated experts is flexibly adjusted. Activating fewer experts reduces inference cost and improves decoding efficiency, while activating more enhances model capability."

Baidu says nothing about how many parameters that leaves. **Baidu has never published an absolute parameter count for ERNIE 5.1** — only the "approximately one-third / approximately one-half" ratio. Any source quoting a hard number is not citing Baidu; this page does not state one.

Distribution is the sharpest contrast with the rest of the Chinese frontier. ERNIE 5.1 has **no published weights**, no Hugging Face repository, no [OpenRouter](https://howaiworks.ai/ai-tools/openrouter) listing, and no Artificial Analysis coverage. It exists as an [API](https://howaiworks.ai/glossary/api) on Baidu's Qianfan platform, a chat box at `ernie.baidu.com`, and a set of leaderboard entries. Baidu's May 13 press release states that "a dedicated cluster powered by Baidu's proprietary Kunlunxin AI chip has supported the training of a key model in the ERNIE 5.1 series" — it stops short of saying ERNIE 5.1 itself was trained on it.

### A note on naming

- **"ERNIE X2" does not exist.** Baidu's reasoning line runs ERNIE X1 → ERNIE X1 Turbo → ERNIE X1.1. No Baidu announcement, press release, or Qianfan model-list entry mentions an X2.
- **Nothing newer than ERNIE 5.1 shipped as of September 6, 2026.** Baidu's ERNIE blog index still lists ERNIE 5.1 (May 9, 2026) as the most recent post, four months on; the ones before it are ERNIE-5.1-Preview (Apr 30), ERNIE-Image (Apr 15), and ERNIE 5.0 (Feb 6). Baidu has shipped no ERNIE 5.2 and no ERNIE 6.
- **"ERNIE," "Wenxin," "文心," and "Yiyan / 文心一言"** all refer to the same family. ERNIE is the model; Wenxin/文心 is the Chinese brand; Yiyan is the consumer chat product.
- The API model ID is lowercase-hyphenated: **`ernie-5.1`**.

## Capabilities

- **Elastic MoE routing**: A variable Top-k router changes how many experts fire per token, so serving cost and capability trade off at [inference](https://howaiworks.ai/glossary/inference) time rather than being fixed at training time.
- **Agentic tool use**: Baidu reports that "on the τ³-bench and SpreadsheetBench-Verified agent evaluation tasks, ERNIE 5.1 surpasses DeepSeek-V4-Pro."
- **Search-grounded answering**: ERNIE 5.1's strongest independently visible result is on arena.ai's *search* leaderboard, where it ranks 5th — the only Chinese model in that top five.
- **Tool-augmented mathematics**: "On AIME26 (with tool use), a challenging mathematical competition benchmark, ERNIE 5.1 scores 99.6 — second only to Gemini 3.1 Pro."
- **Long-form Chinese and English writing**: Baidu's post is subtitled "A Model That Writes Better and Understands You More," and claims creative-writing quality approaching Gemini 3.1 Pro on internal evaluations.
- **128K context with 64K output**: The `ernie-5.1` entry in Baidu's Qianfan model list specifies a 128K [context window](https://howaiworks.ai/glossary/context-window) and a max-output range of `[1, 65536]`.

## Technical Specifications

- **API model ID**: `ernie-5.1`
- **Architecture**: sparse [Mixture-of-Experts](https://howaiworks.ai/glossary/mixture-of-experts) [transformer](https://howaiworks.ai/glossary/transformer) with variable Top-k expert routing
- **Total parameters**: **not published.** Baidu states only "approximately one-third" of ERNIE 5.0's 2.4T.
- **Active parameters**: **not published.** Baidu states only "approximately one-half" of ERNIE 5.0's (itself given only as "<3% activation rate").
- **Context window**: 128K tokens
- **Max output**: 1–65,536 tokens
- **Modality**: text generation (Qianfan classifies `ernie-5.1` under 文本生成, not multimodal)
- **Pre-training**: "Multi-Dimensional Elastic Pre-Training" — elastic depth, elastic width / expert capacity, elastic sparsity
- **Post-training**: four stages — unified SFT, parallel domain-expert training, on-policy [distillation](https://howaiworks.ai/glossary/knowledge-distillation), and general online [reinforcement learning](https://howaiworks.ai/glossary/reinforcement-learning)
- **RL infrastructure**: disaggregated, fully-asynchronous design with FP8 training/inference consistency and heterogeneous resource scheduling
- **Training hardware**: not disclosed for ERNIE 5.1 specifically. Baidu's May 13, 2026 press release says only that "a dedicated cluster powered by Baidu's proprietary Kunlunxin AI chip has supported the training of a key model in the ERNIE 5.1 series."
- **Weights**: not released. **License: proprietary.**

Baidu does not publish a knowledge cutoff, a training-token count, or a parameter count for ERNIE 5.1, so this page does not state one.

## Use Cases

- **Search-grounded assistants**: The one arena.ai board where ERNIE 5.1 is genuinely near the top is search (rank 5), which matches Baidu's core business and its own framing of the model.
- **Chinese-language enterprise deployment**: Qianfan is the practical path for organizations already inside Baidu AI Cloud, and pricing is quoted in CNY for a CNY-billed account.
- **Spreadsheet and workflow agents**: Baidu specifically cites SpreadsheetBench-Verified and τ³-bench as places ERNIE 5.1 beats DeepSeek-V4-Pro — an [agentic-workflow](https://howaiworks.ai/glossary/agentic-workflow) profile rather than a raw-reasoning one.
- **Tool-using math and analysis**: The AIME26 score is reported *with tool use*, so the relevant deployment shape is a model with a calculator/interpreter attached, not a bare chat completion.
- **Long-document processing**: 128K in, up to 64K out, in a single call.
- **Consumer and creative products**: Baidu says the model is "progressively rolled out on over ten creative production agent platforms," including ISEKAI ZERO and Mulan AI.

## Performance / Benchmarks

Baidu's ERNIE 5.1 release post presents its benchmark comparison as an **image**, not as machine-readable text. Only these figures appear in the post's prose, and they are all vendor-reported:

| Claim | Source and exact wording |
|---|---|
| AIME26 (with tool use) | "ERNIE 5.1 scores 99.6 — second only to Gemini 3.1 Pro" |
| τ³-bench, SpreadsheetBench-Verified | "ERNIE 5.1 surpasses DeepSeek-V4-Pro" |
| GPQA, MMLU-Pro | "approaches the performance of leading closed-source models" — **no score given** |
| Arena Search (May 9, 2026) | "ERNIE 5.1 scored 1,223 to claim 4th place globally and 1st among Chinese models" |
| Pre-training cost | "only about 6% of the pre-training cost of comparable models" |

Baidu publishes **no numeric GPQA or MMLU-Pro score**. This page does not state one.

### arena.ai placements (independent, fetched July 8, 2026)

[arena.ai](https://arena.ai/leaderboard/text) — the former LMArena, rebranded in January 2026 — is the only third-party leaderboard carrying ERNIE 5.1. Its text board, in a snapshot dated July 1, 2026, covers 369 models and 7,152,929 votes:

| Leaderboard | Entry | Rank | Score | Votes |
|---|---|---|---|---|
| Text | `ernie-5.1` (Baidu) | **27** | 1468 ±5 | 32,742 (Preliminary) |
| Search | `ernie-5.1` (Baidu) | **5** | 1227 ±10 | 3,829 (Preliminary) |

On the same text board, the Chinese models above ERNIE 5.1 are `qwen3.7-max-preview` (15, 1475), `glm-5.1` (22, 1472), and `glm-5.2 (max)` (26, 1469). Below it: `mimo-v2.5-pro` (29), `kimi-k2.6` (34), `deepseek-v4-pro` (38), `minimax-m3` (54).

**This partly contradicts a live Baidu claim.** Baidu's May 13, 2026 press release states ERNIE 5.1 "ranked #1 among Chinese models on LMArena's text and search leaderboards," and the April 30 preview post claimed "No. 1 among Chinese models and No. 13 globally." **On the text board, neither holds today**: as of the July 1 snapshot, ERNIE 5.1 is the fourth-highest-scoring Chinese model and 27th overall. **On the search board the "#1 among Chinese models" half of the claim does still hold** — no Chinese model outranks it — though the release post's "4th place globally" has slipped to 5th as vote counts grew. Preliminary rankings move; treat launch-day placements as snapshots, not standings.

**Artificial Analysis does not track any Baidu or ERNIE model**, so there is no third-party intelligence-index or independent latency/throughput measurement for ERNIE 5.1. (Artificial Analysis and arena.ai use different scales and are never comparable.)

## Limitations

- **Closed weights, full stop.** No Hugging Face repository, no OpenRouter listing, no self-hosting, no [fine-tuning](https://howaiworks.ai/glossary/fine-tuning) on your own hardware. DeepSeek, Qwen, GLM, Kimi, MiniMax and Tencent all ship downloadable checkpoints for their current flagships; Baidu does not.
- **Baidu open-sources only previous-generation *language* models.** `baidu/ERNIE-4.5-300B-A47B-PT` and the rest of the ERNIE 4.5 family are Apache 2.0 on Hugging Face — first published June 2025, with an `ERNIE-4.5-21B-A3B-Thinking` variant added that September. ERNIE 5.0 and 5.1 weights have never been published. (Outside the LLM line Baidu does ship current-generation weights: `baidu/ERNIE-Image`, Apache 2.0, April 2026.)
- **No published parameter count.** You cannot size a deployment, estimate cost-per-FLOP, or compare efficiency claims against DeepSeek or Qwen, because Baidu gives only a ratio against another model whose active-parameter count is itself given only as "<3%."
- **Every benchmark number is vendor-reported**, and the comparison table is a bitmap. GPQA and MMLU-Pro get an adjective ("approaches the performance of leading closed-source models"), not a score. The 6% pre-training-cost claim has no independent replication and no disclosed accounting method.
- **The vendor's own *text*-leaderboard claim is stale.** "#1 among Chinese models on LMArena's text leaderboard" was true for the preview in late April; on the July 1, 2026 arena.ai snapshot the model is 27th overall and behind three other Chinese entries. (The parallel claim on the *search* board does still hold.)
- **No independent third-party evaluation infrastructure.** Artificial Analysis does not cover it; OpenRouter does not serve it. There is no neutral latency, throughput, or price-performance measurement to check Baidu's efficiency story against.
- **Pricing is CNY-only, and access is China-centric.** Qianfan bills in 元. There is no USD rate card, and the practical onboarding path assumes a Baidu AI Cloud account.
- **On the Hugging Face Hub, Baidu reads as an OCR company, not a frontier lab.** Its most-downloaded asset by a wide margin is [`baidu/Unlimited-OCR`](https://huggingface.co/baidu/Unlimited-OCR) (MIT, ~1.08M downloads in 30 days) — a 3B document-parsing VLM — followed by `baidu/Qianfan-OCR` (~256K). The largest ERNIE 4.5 flagship, `ERNIE-4.5-300B-A47B-PT`, drew about 2,100.

## Pricing & Access

**There is no free tier of weights.** ERNIE 5.1 cannot be self-hosted.

**Hosted access** is sold through Baidu's Qianfan platform. Baidu's [模型服务计费 ("Model Service Billing") page](https://cloud.baidu.com/doc/qianfan/s/wmh4sv6ya), last updated 2026-07-07, lists ERNIE 5.1 under 推理服务 ("inference service") with prompt-length-tiered pricing:

| 子项 (line item) | Translation | 元/千 tokens | equivalent 元/百万 tokens |
|---|---|---|---|
| 输入（输入<=32k） | Input, prompt ≤ 32K | 0.004 | **4** |
| 输出（输入<=32k） | Output, prompt ≤ 32K | 0.018 | **18** |
| 输入（32k<输入<=128k） | Input, prompt 32K–128K | 0.006 | **6** |
| 输出（32k<输入<=128k） | Output, prompt 32K–128K | 0.022 | **22** |

All figures are **yuan (CNY)**. Baidu publishes no USD rate card for ERNIE 5.1, and no conversion is implied here. For reference, the same page prices ERNIE 5.0 higher at every tier (0.006 / 0.024 元/千 tokens at ≤32K), which is consistent with 5.1 being the smaller model.

**Consumer access**: the chat interface at [ernie.baidu.com](https://ernie.baidu.com/), and the Baidu AI Studio playground. Baidu says the model is "progressively rolled out on over ten creative production agent platforms."

**Other ERNIE 5.x model IDs** on Qianfan: `ernie-5.0`, `ernie-5.0-thinking-preview`, `ernie-5.0-thinking-latest`. All share the same 128K context and `[1, 65536]` output range.

## Ecosystem & Tools

- **[ERNIE 5.1 release post](https://ernie.baidu.com/blog/posts/ernie-5.1-0508-release/)** — Baidu's primary announcement, May 9, 2026
- **[Qianfan 模型列表](https://cloud.baidu.com/doc/qianfan/s/rmh4stp0j)** — the authoritative `ernie-5.1` spec: context, max output, category
- **[Qianfan 模型服务计费](https://cloud.baidu.com/doc/qianfan/s/wmh4sv6ya)** — the CNY rate card
- **[ernie.baidu.com](https://ernie.baidu.com/)** — consumer chat, free access to the latest ERNIE model
- **[Baidu on Hugging Face](https://huggingface.co/baidu)** — ERNIE 4.5 (Apache 2.0), ERNIE-Image, Unlimited-OCR. **No ERNIE 5.x.**
- **[baidu/ERNIE-Image](https://huggingface.co/baidu/ERNIE-Image)** — Apache 2.0, 8B single-stream DiT text-to-image model, the current-generation open release
- **Kunlunxin (昆仑芯)** — Baidu's in-house AI accelerator; a Kunlunxin cluster "supported the training of a key model in the ERNIE 5.1 series," per Baidu's May 13 press release

## Community & Resources

- [ERNIE 5.1 Officially Released!](https://ernie.baidu.com/blog/posts/ernie-5.1-0508-release/) — Baidu's official post, May 9, 2026
- [ERNIE-5.1-Preview Tops LMArena Text Leaderboard as No.1 Chinese Model!](https://ernie.baidu.com/blog/posts/ernie-5.1-preview-0430-release-on-lmarena/) — the April 30 preview claim, now superseded
- [ERNIE 5.0: A 2.4 Trillion-Parameter Unified Multimodal Foundation Model](https://ernie.baidu.com/blog/posts/ernie5.0/) — the model ERNIE 5.1 was compressed from
- [Baidu Advances Agent Portfolio…](https://www.prnewswire.com/news-releases/baidu-advances-agent-portfolio-to-embrace-the-agent-era-champions-daily-active-agents-as-key-metric-302771383.html) — Create 2026 press release, May 13, 2026 (Kunlunxin training cluster, DAA metric)
- [Baidu releases Ernie 5.1 with pre-training cost at only 6% of industry peers](https://cntechpost.com/2026/05/09/baidu-releases-ernie-5-1-pre-training-cost-at-only-6-industry-peers/) — CnTechPost
- [百度发布文心 5.1：搜索能力登顶国内，预训练成本仅为业界 6%](https://www.qbitai.com/2026/05/414496.html) — 量子位 (QbitAI)
- [arena.ai text leaderboard](https://arena.ai/leaderboard/text) and [search leaderboard](https://arena.ai/leaderboard/search)
- Compare with [DeepSeek V4](https://howaiworks.ai/models/deepseek), [Qwen3.7-Max](https://howaiworks.ai/models/qwen), [GLM-5.2](https://howaiworks.ai/models/glm), [Kimi K2.6](https://howaiworks.ai/models/kimi), [Tencent Hy3](https://howaiworks.ai/models/hunyuan), and [MiniMax-M3](https://howaiworks.ai/models/minimax-m3)

## Frequently Asked Questions

### When was ERNIE 5.1 released?

May 9, 2026. Baidu's own release post opens: "Today, we officially release ERNIE 5.1." A wider rollout followed at Baidu's Create 2026 developer conference in Beijing on May 13–14.

### Is ERNIE 5.1 open source?

No. The weights are not published. There is no `baidu/ERNIE-5.1` repository on Hugging Face — searching the Hub API for `ERNIE-5` returns only community fine-tunes of ERNIE 4.5 and unrelated repos. arena.ai lists `ernie-5.1` with a proprietary license. Access is API-only through Baidu's Qianfan platform.

### How many parameters does ERNIE 5.1 have?

Baidu has never published an absolute parameter count. The only figure it gives is relative: ERNIE 5.1 "compresses total parameters to approximately one-third and active parameters to approximately one-half" of ERNIE 5.0. Third-party sites that quote a specific number are not citing Baidu — DataLearner, for example, lists "8000亿" (800B) on its ERNIE-5.1-Preview page, a figure equal to ERNIE 5.0's stated 2.4 trillion divided by three. This page does not state a parameter count.

### Is there a model called ERNIE X2?

No. Baidu's reasoning-model line stops at ERNIE X1, ERNIE X1 Turbo, and ERNIE X1.1. No Baidu announcement, Qianfan model-list entry, or press release mentions an "ERNIE X2." Nothing newer than ERNIE 5.1 exists on Baidu's ERNIE blog as of September 6, 2026.

### How much does ERNIE 5.1 cost?

Baidu's Qianfan pricing documentation (last updated 2026-07-07) quotes ERNIE 5.1 in yuan only: 0.004 元/千 tokens input and 0.018 元/千 tokens output for prompts up to 32K, rising to 0.006 / 0.022 元/千 tokens for prompts between 32K and 128K. That is 4 / 18 元 per million tokens at the base tier. Baidu publishes no USD rate card.

### What is the context window of ERNIE 5.1?

128K tokens, with a maximum output range of 1 to 65,536 tokens, per Baidu's Qianfan 模型列表 ("model list") documentation for the `ernie-5.1` model ID.

### Is ERNIE 5.1 multimodal?

Baidu's Qianfan model list places `ernie-5.1` under 文本生成 ("text generation"), not the multimodal categories. Its predecessor ERNIE 5.0 was explicitly pitched as a "Unified Multimodal Foundation Model" spanning text, images, video, and audio; Baidu's ERNIE 5.1 post makes no multimodal claim. Baidu ships image generation separately as ERNIE-Image.

### Why is ERNIE 5.1 cheaper to train?

Baidu attributes it to "Multi-Dimensional Elastic Pre-Training" — elastic depth, elastic width/expert capacity, and elastic sparsity. On the last of these it writes: "Through a variable Top-k routing mechanism, the number of activated experts is flexibly adjusted." Baidu says this yields "leading foundational performance at its model scale using only about 6% of the pre-training cost of comparable models." This is a vendor claim with no independent replication.

### Where does ERNIE 5.1 rank on the arena.ai leaderboards?

On the arena.ai text leaderboard (snapshot dated July 1, 2026, 369 models, 7.15M votes), `ernie-5.1` sits at rank 27 with a score of 1468±5 from 32,742 votes, still flagged Preliminary. On the arena.ai search leaderboard it sits at rank 5 with 1227±10 from 3,829 votes.

### Which ERNIE weights can I actually download?

The ERNIE 4.5 family — including `baidu/ERNIE-4.5-300B-A47B-PT` and `baidu/ERNIE-4.5-VL-424B-A47B-PT` — is Apache 2.0 on Hugging Face, as is the newer `baidu/ERNIE-Image`. No ERNIE 5.x language-model weights have ever been released.

## Related

### Related models

- [DeepSeek V4](https://howaiworks.ai/models/deepseek)
- [Qwen3.8-Max](https://howaiworks.ai/models/qwen)
- [GLM-5.3](https://howaiworks.ai/models/glm)
- [Kimi K3](https://howaiworks.ai/models/kimi)
- [Tencent Hy4 preview](https://howaiworks.ai/models/hunyuan)
- [MiniMax-M3](https://howaiworks.ai/models/minimax-m3)

### Related tools

- [OpenRouter](https://howaiworks.ai/ai-tools/openrouter)

---

Source: https://howaiworks.ai/models/ernie — HowAIWorks.ai