// head_to_head

DeepSeek V4 Flash Vision Exp vs Qwen3.8 Omni Flash

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.

OpenRouter + LiveBenchAll comparisonsFull leaderboard

Qwen3.8 Omni Flash is the cheaper of the two; neither can be ranked on quality here.

Qwen3.8 Omni Flash 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.

deepseek

DeepSeek V4 Flash Vision Exp

Blended / 1M
$0.330
Context
1.0M
Released
Aug 21, 2026
Overall score
76.8
reasoningtool callingimage inputprompt caching

qwen

Qwen3.8 Omni Flash

Blended / 1M
$0.230
Context
1M
Released
Sep 21, 2026
Overall score
Not evaluated
reasoningtool callingimage inputaudio inputvideo inputprompt caching

Specs and pricing

MetricDeepSeek V4 Flash Vision ExpQwen3.8 Omni Flash
LiveBench overall

Mean of the seven LiveBench category scores, 0–100. Higher is better.

76.8
Cost per point

Measured benchmark spend divided by overall score — dollars per point of capability.

$0.0277
Blended price / 1M

3:1 input:output mix, the usual shape of production traffic.

$0.330$0.230win
Input price / 1M$0.220$0.150win
Output price / 1M$0.660$0.470win
Cached input / 1M

Price of an input token served from the prompt cache, where the provider publishes one.

$0.0070win$0.016
Context window1.0M1M
Max output tokens944Kwin131K

Benchmarks by category

An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — DeepSeek V4 Flash Vision Exp on top, Qwen3.8 Omni Flash below, both out of 100.

Agentic coding
65.1
Coding
68.2
Reasoning
85.4
Mathematics
87.8
Data analysis
79.5
Language
80.4
Instruction following
71.0

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.

WorkloadDeepSeek V4 Flash Vision ExpQwen3.8 Omni Flash
Support chatbot

1.2K in / 400 out × 200K requests

$90.26/mo$63.95/mo
RAG assistant

8K in / 600 out × 100K requests

$130.40/mo$94.60/mo
Coding agent

40K in / 4K out × 20K requests

$109.52/mo$82.56/mo
Document extraction

20K in / 1.5K out × 50K requests

$258.85/mo$178.55/mo
Bulk classification

500 in / 20 out × 5M requests

$509.50/mo$355.00/mo
Run these two through the cost calculator

Which should you pick?

You are cost-constrained

Qwen3.8 Omni Flash

Cheaper on blended list price at $0.230 per million tokens.

DeepSeek V4 Flash Vision Exp vs Qwen3.8 Omni Flash FAQ

Which is better, DeepSeek V4 Flash Vision Exp or Qwen3.8 Omni Flash?

Qwen3.8 Omni Flash is the cheaper of the two; neither can be ranked on quality here. Qwen3.8 Omni Flash 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 DeepSeek V4 Flash Vision Exp cheaper than Qwen3.8 Omni Flash?

Qwen3.8 Omni Flash is cheaper. On a 3:1 input:output blend, DeepSeek V4 Flash Vision Exp lists at $0.330 per million tokens and Qwen3.8 Omni Flash at $0.230 — Qwen3.8 Omni Flash is 43% cheaper. Input and output are priced separately — DeepSeek V4 Flash Vision Exp charges $0.220 in and $0.660 out, Qwen3.8 Omni Flash charges $0.150 and $0.470 — so the model that looks cheaper flips depending on how output-heavy your workload is.

Does DeepSeek V4 Flash Vision Exp or Qwen3.8 Omni Flash have a bigger context window?

They are effectively the same — 1.0M for DeepSeek V4 Flash Vision Exp and 1M for Qwen3.8 Omni Flash.

Do DeepSeek V4 Flash Vision Exp and Qwen3.8 Omni Flash support prompt caching?

Both publish a cached-input rate: $0.0070 per million for DeepSeek V4 Flash Vision Exp and $0.016 for Qwen3.8 Omni Flash, against full input rates of $0.220 and $0.150. 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.
  • ScoresLiveBench release 2026-06-25, 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.