Compare / head-to-head
ERNIE 5.0vs
Qwen3 30B-A3B Instruct 2507
ERNIE 5.0 leads 5 of 5 shared benchmarks. Qwen3 30B-A3B Instruct 2507 has the larger context window (262K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
5 – 0
ERNIE 5.0 leads
Cheaper per token
—
list price, input + output
Larger context
Qwen3 30B-A3B Instruct 2507
262K tokens
Providers
2 providers
Baidu · Qwen
ERNIE 5.0
Baidu · released 2026-02-06
textvision
- Context
- 128K
- Max out
- 66K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 22 · 6 core
Qwen3 30B-A3B Instruct 2507
Qwen · released 2025-07-30
text
Quality
Benchmark matrix
5 shared · 17 only ERNIE 5.0 · 14 only Qwen3 30B-A3B Instruct 2507| Benchmark | ERNIE 5.0 | Qwen3 30B-A3B Instruct 2507 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 5 | ||||
| GPQA Diamond | 86.36% ↗ | 55.6% ↗epoch run | +30.8 pt | ERNIE 5.0 |
| MMLU-Pro | 83.8% ↗ | 78.4% ↗ | +5.4 pt | ERNIE 5.0 |
| LMArena Elo | 1446.4 ↗ | 1382.7 ↗ | +63.7 | ERNIE 5.0 |
| IFEval | 93.35% ↗ | 84.7% ↗ | +8.6 pt | ERNIE 5.0 |
| ZebraLogic | 96.5% ↗ | 90% ↗ | +6.5 pt | ERNIE 5.0 |
| Only ERNIE 5.0 reports · 17 | ||||
| AIME 2025 | 89.06% ↗ | not reported | — | — |
| ACEBench Chinese | 89.6% ↗ | not reported | — | — |
| ACEBench English | 87.7% ↗ | not reported | — | — |
| BBEH | 66.63% ↗ | not reported | — | — |
| BFCL v4 | 66.47% ↗ | not reported | — | — |
| BrowseComp-ZH | 64.71% ↗ | not reported | — | — |
| ChineseSimpleQA | 86.03% ↗ | not reported | — | — |
| HMMT 2025 | 79.58% ↗ | not reported | — | — |
| HumanEval+ | 94.48% ↗ | not reported | — | — |
| Humanity's Last Exam | 25.81% ↗ | not reported | — | — |
| LiveCodeBench v6 | 76.21% ↗ | not reported | — | — |
| MBPP+ | 82.54% ↗ | not reported | — | — |
| Multi-IF | 85.56% ↗ | not reported | — | — |
| MultiChallenge | 65.98% ↗ | not reported | — | — |
| SimpleQA | 74.01% ↗ | not reported | — | — |
| SpreadsheetBench | 40.08% ↗ | not reported | — | — |
| Tau2-Bench | 78.79% ↗ | not reported | — | — |
| Only Qwen3 30B-A3B Instruct 2507 reports · 14 | ||||
| Aider-Polyglot | not reported | 35.6% ↗ | — | — |
| AIME25 | not reported | 61.3% ↗ | — | — |
| Arena-Hard v2 | not reported | 69% ↗ | — | — |
| BFCL-v3 | not reported | 65.1% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 2% ↗epoch run | — | — |
| GPQA | not reported | 70.4% ↗ | — | — |
| HMMT25 | not reported | 43% ↗ | — | — |
| LiveBench 20241125 | not reported | 69% ↗ | — | — |
| LiveCodeBench v6 (25.02-25.05) | not reported | 43.2% ↗ | — | — |
| MMLU-Redux | not reported | 89.3% ↗ | — | — |
| MultiPL-E | not reported | 83.8% ↗ | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 62.2% ↗epoch run | — | — |
| SuperGPQA | not reported | 53.4% ↗ | — | — |
| TAU2-Retail | not reported | 57% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is ERNIE 5.0 minus Qwen3 30B-A3B Instruct 2507 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | ERNIE 5.0 | Qwen3 30B-A3B Instruct 2507 | Edge |
|---|---|---|---|
| Context window | 128K tokens | 262K tokens | Qwen3 30B-A3B Instruct 2507 |
| Max output | 66K tokens | — | — |
| Input price / 1M | — | $0.2 ↗ | — |
| Output price / 1M | — | $0.8 ↗ | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | $1 | — |
| Modalities | text · vision | text | ERNIE 5.0 |
| Released | 2026-02-06 | 2025-07-30 | — |
| Cited benchmark scores | 22 | 19 | — |
Reliability
Provider status
Live from /statusMore matchups
ERNIE 5.0 vs …
Models sharing the most benchmarksMore matchups
Qwen3 30B-A3B Instruct 2507 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run ERNIE 5.0 and Qwen3 30B-A3B Instruct 2507 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.