// head_to_head
ERNIE-Light-DT-1.0 vs DeepSeek V4.1 Flash 1017
Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.
DeepSeek V4.1 Flash 1017 is the cheaper of the two; neither can be ranked on quality here.
ERNIE-Light-DT-1.0 does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.
baidu
ERNIE-Light-DT-1.0
- Blended / 1M
- $1.83
- Context
- 1.0M
- Released
- Nov 10, 2026
- Overall score
- Not evaluated
deepseek
DeepSeek V4.1 Flash 1017
- Blended / 1M
- $0.130
- Context
- 1.0M
- Released
- Oct 17, 2026
- Overall score
- 70.3
Specs and pricing
| Metric | ERNIE-Light-DT-1.0 | DeepSeek V4.1 Flash 1017 |
|---|---|---|
| LiveBench overall Weighted mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 70.3 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.0019 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $1.83 | $0.130win |
| Input price / 1M | $1.13 | $0.0060win |
| Output price / 1M | $3.94 | $0.500win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.113 | $0.0050win |
| Context window | 1.0M | 1.0M |
| Max output tokens | 131K | 944Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — ERNIE-Light-DT-1.0 on top, DeepSeek V4.1 Flash 1017 below, both out of 100.
What each one costs to run
Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.
| Workload | ERNIE-Light-DT-1.0 | DeepSeek V4.1 Flash 1017 |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $513.18/mo | $41.37/mo |
| RAG assistant 8K in / 600 out × 100K requests | $733.60/mo | $34.40/mo |
| Coding agent 40K in / 4K out × 20K requests | $649.68/mo | $44.24/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $1,375/mo | $43.45/mo |
| Bulk classification 500 in / 20 out × 5M requests | $2,711/mo | $64.50/mo |
Which should you pick?
You are cost-constrained
DeepSeek V4.1 Flash 1017
Cheaper on blended list price at $0.130 per million tokens.
ERNIE-Light-DT-1.0 vs DeepSeek V4.1 Flash 1017 FAQ
Which is better, ERNIE-Light-DT-1.0 or DeepSeek V4.1 Flash 1017?
DeepSeek V4.1 Flash 1017 is the cheaper of the two; neither can be ranked on quality here. ERNIE-Light-DT-1.0 does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.
Is ERNIE-Light-DT-1.0 cheaper than DeepSeek V4.1 Flash 1017?
DeepSeek V4.1 Flash 1017 is cheaper. On a 3:1 input:output blend, ERNIE-Light-DT-1.0 lists at $1.83 per million tokens and DeepSeek V4.1 Flash 1017 at $0.130 — DeepSeek V4.1 Flash 1017 is 14× cheaper. Input and output are priced separately — ERNIE-Light-DT-1.0 charges $1.13 in and $3.94 out, DeepSeek V4.1 Flash 1017 charges $0.0060 and $0.500 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Does ERNIE-Light-DT-1.0 or DeepSeek V4.1 Flash 1017 have a bigger context window?
They are effectively the same — 1.0M for ERNIE-Light-DT-1.0 and 1.0M for DeepSeek V4.1 Flash 1017.
Do ERNIE-Light-DT-1.0 and DeepSeek V4.1 Flash 1017 support prompt caching?
Both publish a cached-input rate: $0.113 per million for ERNIE-Light-DT-1.0 and $0.0050 for DeepSeek V4.1 Flash 1017, against full input rates of $1.13 and $0.0060. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.
Related comparisons
How these numbers are produced
- Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
- Scores — LiveBench release 2026-11-06, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
- Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
- “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.
Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.