Model comparison
DeepSeek V4 Flash vs DeepSeek V4 Pro
Two DeepSeek tiers compared on the figures that decide which one a workload actually needs.
Recorded profile verification October 1, 2026Individual provider fields may changeEvidence confidence: Insufficient data — see receipts below
DeepSeek
DeepSeek V4 Flash
Balanced · Open weights
DeepSeek
DeepSeek V4 Pro
Frontier · Open weights
| Specification | DeepSeek V4 Flash | DeepSeek V4 Pro |
|---|---|---|
| Provider | ||
| Provider | DeepSeek | DeepSeek |
| Tier | ||
| Tier | Balanced | Frontier |
| Context window | ||
| Context window | 1M | 1M |
| Max output | ||
| Max output | Not verifiedUnverified | 384K |
| Input / 1M tokens | ||
| Input / 1M tokens | Not verifiedUnverified | $1.32 |
| Output / 1M tokens | ||
| Output / 1M tokens | Not verifiedUnverified | $3.96 |
| Weights | ||
| Weights | Open | Open |
| Parameters | ||
| Parameters | 284B total / 13B active (MoE) | 1.6T total / 49B active (MoE) |
| Reasoning levels | ||
| Reasoning levels | low, high, max | low, high, max |
| Modalities | ||
| Modalities | text | text |
| License | ||
| License | MIT | MIT |
| API model id | ||
| API model id | deepseek-v4-flash | deepseek-v4-pro |
| Released | ||
| Released | April 24, 2026 | April 24, 2026 |
| SWE-bench Verified · source date DeepSeek V4 Flash: 2026-04-24 | ||
| SWE-bench Verified · source date DeepSeek V4 Flash: 2026-04-24 | 79 | Not verifiedUnverified |
| SWE-bench Verified [max] · source date DeepSeek V4 Pro: 2026-04-24 | ||
| SWE-bench Verified [max] · source date DeepSeek V4 Pro: 2026-04-24 | Not verifiedUnverified | 80.6 |
| GPQA Diamond · source date DeepSeek V4 Flash: 2026-04-24 | ||
| GPQA Diamond · source date DeepSeek V4 Flash: 2026-04-24 | 88.1 | Not verifiedUnverified |
| GPQA Diamond [max] · source date DeepSeek V4 Pro: 2026-04-24 | ||
| GPQA Diamond [max] · source date DeepSeek V4 Pro: 2026-04-24 | Not verifiedUnverified | 90.1 |
| Humanity's Last Exam · source date DeepSeek V4 Flash: 2026-04-24 | ||
| Humanity's Last Exam · source date DeepSeek V4 Flash: 2026-04-24 | 45.1 | Not verifiedUnverified |
| Humanity's Last Exam [max] · source date DeepSeek V4 Pro: 2026-04-24 | ||
| Humanity's Last Exam [max] · source date DeepSeek V4 Pro: 2026-04-24 | Not verifiedUnverified | 48.2 |
| Humanity's Last Exam · source date DeepSeek V4 Pro: 2026-08-13 | ||
| Humanity's Last Exam · source date DeepSeek V4 Pro: 2026-08-13 | Not verifiedUnverified | 42.7 |
| Humanity's Last Exam · source date DeepSeek V4 Pro: 2026-08-13 | ||
| Humanity's Last Exam · source date DeepSeek V4 Pro: 2026-08-13 | Not verifiedUnverified | 60 |
| MMLU-Pro · source date DeepSeek V4 Flash: 2026-04-24 | ||
| MMLU-Pro · source date DeepSeek V4 Flash: 2026-04-24 | 86.2 | Not verifiedUnverified |
| MMLU-Pro [max] · source date DeepSeek V4 Pro: 2026-04-24 | ||
| MMLU-Pro [max] · source date DeepSeek V4 Pro: 2026-04-24 | Not verifiedUnverified | 87.5 |
| Artificial Analysis Intelligence Index [max] · source date DeepSeek V4 Pro: 2026-10-01 | ||
| Artificial Analysis Intelligence Index [max] · source date DeepSeek V4 Pro: 2026-10-01 | Not verifiedUnverified | 36 |
| Terminal-Bench 2.1 · source date DeepSeek V4 Pro: 2026-08-13 | ||
| Terminal-Bench 2.1 · source date DeepSeek V4 Pro: 2026-08-13 | Not verifiedUnverified | 87.9 |
| DeepSWE 1.1 · source date DeepSeek V4 Pro: 2026-08-13 | ||
| DeepSWE 1.1 · source date DeepSeek V4 Pro: 2026-08-13 | Not verifiedUnverified | 62.7 |
Dates beside benchmark scores refer to the linked source’s publication or capture date unless that source explicitly states when it measured the result. A shared URL alone does not establish comparable runs; unverified setups are marked. Prices are USD per million tokens at standard rates, excluding batch and caching discounts. Values we could not confirm from the provider are shown as “Not verified” rather than estimated.
Benchmark receipts
Where each score comes from, and how far it can be compared across models.
- Source date2026-10-01Artificial Analysis (DeepSeek V4 Pro 0813 (max))
Composite index blending reasoning, knowledge, and coding evals into one 0–100 score.
Comparable with caveatOnly same-version scores are comparable. v4.3.2 was rebaselined (the top score fell from 66 on v4.1.1 to ~58), so v4.1.1 figures must not be compared with v4.3.2 figures — check the cited source and its date.
- Source date2026-08-13DeepSeek August 13 GA changelog (vendor) (0813 GA refresh; vendor report, exact run configuration unverified)
Agentic software engineering: pass@1 on real repo tasks with reported cost per completed task.
Directly comparable
- Source date2026-08-13DeepSeek August 13 GA changelog (vendor) (0813 GA refresh; vendor report, exact run configuration unverified)
Agentic terminal work: multi-step tasks executed in a sandboxed shell environment.
Comparable with caveatNot comparable with Terminal-Bench 3.0 or 4.0 (different task sets) or v1; harness configuration (container, time limits) also shifts results.
- Source date2026-08-13DeepSeek August 13 GA changelog (vendor) (0813 GA refresh; with tools; vendor report, exact run configuration unverified)
Frontier-knowledge ceiling: extremely hard multi-domain questions written to be near-impossible without deep expertise.
Comparable with caveatSubset/tool configurations (text-only vs with-tools) differ between vendors and materially change scores; must match configuration to compare.
- Source date2026-08-13DeepSeek August 13 GA changelog (vendor) (0813 GA refresh; without tools; vendor report, exact run configuration unverified)
Frontier-knowledge ceiling: extremely hard multi-domain questions written to be near-impossible without deep expertise.
Comparable with caveatSubset/tool configurations (text-only vs with-tools) differ between vendors and materially change scores; must match configuration to compare.
- Source date2026-04-24DeepSeek V4 Pro HF model card (vendor, Think Max, exact match) (Original April 24 checkpoint; historical vendor evaluation, not the served 0813 refresh)
Broad knowledge across 14 domains with harder, reasoning-focused multiple-choice questions.
Comparable with caveatFrontier models cluster in the high 80s–90s; near-ceiling differences are weak evidence. CoT vs direct answering must match.
- Source date2026-04-24DeepSeek V4 Pro HF model card (vendor, Think Max, with tools; 37.7 no tools) (Original April 24 checkpoint; historical vendor evaluation, not the served 0813 refresh)
Frontier-knowledge ceiling: extremely hard multi-domain questions written to be near-impossible without deep expertise.
Comparable with caveatSubset/tool configurations (text-only vs with-tools) differ between vendors and materially change scores; must match configuration to compare.
- Source date2026-04-24DeepSeek V4 Pro HF model card (vendor, Think Max, pass@1) (Original April 24 checkpoint; historical vendor evaluation, not the served 0813 refresh)
Graduate-level science reasoning: multiple-choice questions written by domain PhDs.
Comparable with caveatDo not treat a fixed two-point band as universal measurement uncertainty; it depends on run count and evaluation setup. Chain-of-thought, prompt, and tool policy must match. The small pool limits fine-grained conclusions, and frontier scores can be near ceiling; Artificial Analysis dropped it from its Intelligence Index in v4.3.
- Source date2026-04-24DeepSeek original V4 Pro HF card (vendor, Think Max; verifier/harness comparability unverified) (Original April 24 checkpoint; historical vendor evaluation, not the served 0813 refresh)
Real GitHub issue resolution: does the model's patch pass the hidden tests.
Comparable with caveatPost-audit vendor claims and pre-audit scores sit on different task trust levels; scaffolding (agent harness, compute budget) also dominates results. Never aggregate across scaffolds.
- Source date2026-04-24DeepSeek family HF model card (vendor, Think Max, exact match)
Broad knowledge across 14 domains with harder, reasoning-focused multiple-choice questions.
Comparable with caveatFrontier models cluster in the high 80s–90s; near-ceiling differences are weak evidence. CoT vs direct answering must match.
- Source date2026-04-24DeepSeek family HF model card (vendor, Think Max, with tools; 34.8 no tools)
Frontier-knowledge ceiling: extremely hard multi-domain questions written to be near-impossible without deep expertise.
Comparable with caveatSubset/tool configurations (text-only vs with-tools) differ between vendors and materially change scores; must match configuration to compare.
- Source date2026-04-24DeepSeek family HF model card (vendor, Think Max, pass@1)
Graduate-level science reasoning: multiple-choice questions written by domain PhDs.
Comparable with caveatDo not treat a fixed two-point band as universal measurement uncertainty; it depends on run count and evaluation setup. Chain-of-thought, prompt, and tool policy must match. The small pool limits fine-grained conclusions, and frontier scores can be near ceiling; Artificial Analysis dropped it from its Intelligence Index in v4.3.
- Source date2026-04-24DeepSeek original family HF model card (vendor, Think Max; historical April checkpoint, not the 0731 refresh)
Real GitHub issue resolution: does the model's patch pass the hidden tests.
Comparable with caveatPost-audit vendor claims and pre-audit scores sit on different task trust levels; scaffolding (agent harness, compute budget) also dominates results. Never aggregate across scaffolds.
Pricing tiers
DeepSeek V4 Flash: Retired 2026-09-10 — no longer sold. Historical checkpoint tariffs are not verified here. The legacy deepseek-v4-flash ID temporarily routes to V4.1-Flash at $0.30/$1.20 peak or $0.15/$0.60 off-peak. MIT weights remain. A current 384K API output limit belongs to the routed V4.1-Flash; the original checkpoint's hard output cap is not verified.
DeepSeek V4 Pro: Official peak rates $1.32/$3.96 per MTok; off-peak $0.66/$1.98 (cache-miss). Cache hit $0.044 peak / $0.022 off-peak. Peak hours are weekdays 01:00–04:00 and 06:00–10:00 UTC, excluding Chinese public holidays; weekends, holidays and all other hours are off-peak. Open weights (MIT). 1.6T/49B active MoE.
DeepSeek V4 Flash
DeepSeek V4 Flash was the efficient open-weight (MIT) sibling of V4 Pro — retired from DeepSeek's API on 2026-09-10 and replaced by V4.1-Flash.
Best for
- Open-weight deployments
- Cost-sensitive inference
- High-volume serving
Watch out
Retired from the first-party API: the legacy ID reaches V4.1-Flash, not these original weights. MIT weights remain for self-hosting. Do not transfer the current alias's output cap or vision capabilities to the original text-only checkpoint.
DeepSeek V4 Pro
DeepSeek V4 Pro is DeepSeek's large open-weight (MIT) MoE with a 1M-token context — DeepSeek says the cheaper V4.1-Flash now beats it.
Best for
- Cost-sensitive hosted agents
- Open-weight deployments
- High-volume coding
Watch out
The current tariff identifies DeepSeek-V4-Pro-0813. DeepSeek's current changelog reverses the earlier September 14 re-route plan and says it will continue serving V4 Pro at unchanged rates until further notice. Original April benchmark/checkpoint results are not the current 0813 evaluation. Self-hosting memory depends on quantization, offloading and hardware.
When the cheaper one wins
One of these rows is missing a verified output price, so this page will not name a cheaper winner. Run the model picker against the actual job instead of guessing.
Run the model pickerEvidence confidence: Insufficient data
How strong and complete the evidence behind this comparison is — not a prediction of which model is better. Hard limitations apply: missing pricing; evidence too sparse.
- Pricing incomplete — No verified token pricing for DeepSeek V4 Flash.
- 4/5 core specs verified on both sides — Not published for at least one side: max output.
- No directly aligned benchmark setup — No benchmark has matching cited sources, source dates, units, effort and recorded run details on both models.
- Verified within the last 90 days — Older of the two catalog checks was 25 days ago.
- Both models carry source citations — Each side has at least two catalog sources on record.
Source receipts
Each catalog figure was checked against the provider or an independent second source on the date shown.
- DeepSeek V4 Flash: Whole-profile primary source check October 1 (accessed 2026-10-01)
- DeepSeek V4 Flash: Whole-profile primary source check October 1 (accessed 2026-10-01)
- DeepSeek V4 Flash: DeepSeek — V4 news (accessed 2026-08-29)
- DeepSeek V4 Flash: DeepSeek API pricing (accessed 2026-08-29)
- DeepSeek V4 Flash: DeepSeek — V4.1-Flash release (V4-Flash retired) (accessed 2026-09-26)
- DeepSeek V4 Flash: DeepSeek API pricing (legacy id billed at V4.1-Flash rates) (accessed 2026-09-26)
- DeepSeek V4 Pro: Whole-profile primary source check October 1 (accessed 2026-10-01)
- DeepSeek V4 Pro: Whole-profile primary source check October 1 (accessed 2026-10-01)
- DeepSeek V4 Pro: DeepSeek current API version, holiday tariff and 0813 GA benchmark publication (accessed 2026-10-01)
- DeepSeek V4 Pro: DeepSeek current tariff and served endpoint versions (accessed 2026-10-01)
- DeepSeek V4 Pro: DeepSeek — V4 news (accessed 2026-08-29)
- DeepSeek V4 Pro: DeepSeek API pricing (accessed 2026-08-29)
- DeepSeek V4 Pro: DeepSeek — V4.1-Flash release (V4-Pro phase-out) (accessed 2026-09-26)
- DeepSeek V4 Pro: DeepSeek API changelog (accessed 2026-09-26)
Related comparisons
- Amazon Nova 2 Lite vs DeepSeek V4 Flash
- Amazon Nova 2 Pro vs DeepSeek V4 Flash
- Base 1 vs DeepSeek V4 Flash
- Claude Fable 5.1 vs DeepSeek V4 Pro
- Claude Fable 5 vs DeepSeek V4 Pro
- Claude Mythos 5.1 vs DeepSeek V4 Pro
Diving deeper on one model? DeepSeek V4 Flash · DeepSeek V4 Pro
Common questions
DeepSeek V4 Flash vs DeepSeek V4 Pro
Answered from the verified figures on this page rather than general guidance.
Which has the larger context window, DeepSeek V4 Flash or DeepSeek V4 Pro?
Do DeepSeek V4 Flash and DeepSeek V4 Pro support the same reasoning levels?
Should I use DeepSeek V4 Flash or DeepSeek V4 Pro?
Next step
Choosing between them
The verified figures that separate this pair, computed from the catalog rather than restated boilerplate.
- Positioning: DeepSeek V4 Flash sits in the balanced tier, DeepSeek V4 Pro in the frontier tier. Compare accuracy, reliability, latency, and cost on representative tasks before choosing.
Choosing a harness rather than a model — Cursor, Copilot, Claude Code, Windsurf, or Lovable? Compare agentic harnesses · Latest releases.