DeepSeek vs DeepSeek · synced 2026-08-11

deepseek-v3.2 vs deepseek-v4-flash

deepseek-v4-flash is 2.0× cheaper on input than deepseek-v3.2.

deepseek-v3.2 deepseek-v4-flash
Input / 1M $0.28 $0.14
Cached input / 1M $0.003
Output / 1M $0.40 $0.28
Context window 164K 1M
Max output 164K 8K
Provider DeepSeek DeepSeek
Vision No No
Tool calling Yes Yes
Token count estimated estimated

Capability for the money

Independent measurement · Artificial Analysis · read 2026-08-11

deepseek-v4-flash is 1.8× cheaper than deepseek-v3.2 and scores higher on hard reasoning (GPQA Diamond) — there is no case for paying more here.

  1. Hard reasoning

    graduate-level science and multi-step analysis

    GPQA Diamond
    deepseek-v3.275.1%
    deepseek-v4-flash90.8%

    deepseek-v4-flash +15.7

Averaged over the benchmarks both models sat, deepseek-v4-flash returns 518.9 points of pass rate per dollar of blended price against 242.3 — the ratio that decides whether the dearer model earns its rate.

Each figure is a pass rate: the share of that test set the model answered correctly, measured by a third party rather than self-reported by the lab. Rows are different tests of different difficulty, so compare a model against the other model, not one row against the next. No published following instructions, coding or tool use score for one of these models, so those lenses are omitted. And a benchmark is not your workload — it narrows the shortlist, it does not pick for you.

Which to pick

List price is not production cost. Output length, cache hit rate, retries and task quality move the real number more than the headline rate does — price your own prompt on the token ledger before deciding.

Related comparisons