Model comparison
Claude Opus 5 vs DeepSeek V4 Flash
Anthropic against DeepSeek, compared on context, price, and attributed benchmark results.
Claude Opus 5: profile checked October 25, 2026DeepSeek V4 Flash: profile checked October 25, 2026Individual provider fields may changeEvidence confidence: Insufficient data — see receipts below
Anthropic
Claude Opus 5
Frontier
DeepSeek
DeepSeek V4 Flash
Balanced · Open weights
| Specification | Claude Opus 5 | DeepSeek V4 Flash |
|---|---|---|
| Provider | ||
| Provider | Anthropic | DeepSeek |
| Tier | ||
| Tier | Frontier | Balanced |
| Context window | ||
| Context window | 1M | 1M |
| Max output | ||
| Max output | 128K | Not verifiedUnverified |
| Input / 1M tokens | ||
| Input / 1M tokens | $5 | Not verifiedUnverified |
| Output / 1M tokens | ||
| Output / 1M tokens | $25 | Not verifiedUnverified |
| Weights | ||
| Weights | Closed | Open |
| Parameters | ||
| Parameters | Not verifiedUnverified | 284B total / 13B active (MoE) |
| Reasoning levels | ||
| Reasoning levels | low, medium, high, xhigh, max | low, high, max |
| Modalities | ||
| Modalities | text, image | text |
| License | ||
| License | Not verifiedUnverified | MIT |
| API model id | ||
| API model id | claude-opus-5 | deepseek-v4-flash |
| Released | ||
| Released | July 24, 2026 | April 24, 2026 |
| Humanity's Last Exam · source date Claude Opus 5: 2026-09-01, DeepSeek V4 Flash: 2026-04-24 (comparability not verified) | ||
| Humanity's Last Exam · source date Claude Opus 5: 2026-09-01, DeepSeek V4 Flash: 2026-04-24 (comparability not verified) | 63.6 | 45.1 |
| Terminal-Bench 2.1 · source date Claude Opus 5: 2026-09-02 | ||
| Terminal-Bench 2.1 · source date Claude Opus 5: 2026-09-02 | 89.1 | Not verifiedUnverified |
| SWE-bench Verified · source date DeepSeek V4 Flash: 2026-04-24 | ||
| SWE-bench Verified · source date DeepSeek V4 Flash: 2026-04-24 | Not verifiedUnverified | 79 |
| GPQA Diamond · source date DeepSeek V4 Flash: 2026-04-24 | ||
| GPQA Diamond · source date DeepSeek V4 Flash: 2026-04-24 | Not verifiedUnverified | 88.1 |
| MMLU-Pro · source date DeepSeek V4 Flash: 2026-04-24 | ||
| MMLU-Pro · source date DeepSeek V4 Flash: 2026-04-24 | Not verifiedUnverified | 86.2 |
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 date2026-09-02Google Gemini 3.8 Flash evaluation table quoting public leaderboard / Artificial Analysis, Terminus 2; not a Google-run Opus evaluation (Cross-provider quotation; common effort and experiment date 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 date2026-09-01Anthropic Fable 5.1 comparison table (vendor, with tools; 56.6 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, 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
Claude Opus 5: Standard USD/MTok input/output $5/$25 with no long-context surcharge; cache reads $0.50, 5-minute writes $6.25, 1-hour writes $10. Batch $2.50/$12.50. First-party Fast research preview $10/$50, unavailable with Batch; caching and data-residency modifiers stack. US-only inference adds 10% to input/output and cache reads/writes on Anthropic-operated routes where supported; Foundry uses US Data Zone Standard. Bedrock and Google-managed Google Cloud have independent regional prices. Batch 300K output requires output-300k-2026-03-24; standard stays 128K. Knowledge cutoff May 2026; retirement not sooner than July 24, 2027. Server-side tools can add usage fees; tool schemas/results and thinking consume billable tokens.
DeepSeek V4 Flash: Retired 2026-09-10 — no longer sold as the original checkpoint. Historical checkpoint tariffs are not verified here. The legacy deepseek-v4-flash ID now routes to V4.1-Flash-1017 at $0.30/$1.20 peak or $0.15/$0.60 off-peak, with $0.006/$0.003 cache reads. MIT weights remain. The 384K current API output ceiling and vision support belong to the routed revision, not these original text-only weights.
Claude Opus 5
Claude Opus 5 is Anthropic's July 2026 agentic-coding model, now 'Active (legacy)'. Its successor Opus 5.5 has lower base input/output prices ($4/$20 vs $5/$25 per million tokens); performance depends on the workload and evaluation settings.
Best for
- Complex agentic coding
- Enterprise work
- Long-context analysis
Watch out
Active legacy, superseded by Opus 5.5. On Opus 5, xhigh/max rejects disabled thinking; use a large output budget because thinking shares max_tokens. Test migration: Opus 5.5 makes adaptive thinking always on, rejects forced tool use and the older computer-use tool, and changes thinking-block/progress behavior. Partner cloud lifecycles and prices can differ.
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
The original V4 Flash API checkpoint retired September 10; its old ID is still accepted and now serves Flash-1017. Original MIT weights remain available for self-hosting. Original API low/high/max settings are historical controls, not evidence of a matching native hosted menu or successor performance.
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.
- 3/5 core specs verified on both sides — Not published for at least one side: max output, parameter count.
- 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 1 day 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.
- Claude Opus 5: Whole-record source check: exact model card (accessed 2026-10-25)
- Claude Opus 5: Whole-record source check: current publisher pricing, controls or access terms (accessed 2026-10-25)
- Claude Opus 5: Whole-record source check: current publisher pricing, controls or access terms (accessed 2026-10-25)
- Claude Opus 5: Whole-record source check: current publisher pricing, controls or access terms (accessed 2026-10-25)
- Claude Opus 5: Anthropic news — Claude Opus 5 (accessed 2026-08-29)
- Claude Opus 5: Anthropic — Claude Opus 5.5 (Opus 5 successor) (accessed 2026-09-26)
- DeepSeek V4 Flash: Whole-record check: current publisher specifications, licence, access or tariff (accessed 2026-10-25)
- DeepSeek V4 Flash: Whole-record check: current publisher specifications, licence, access or tariff (accessed 2026-10-25)
- DeepSeek V4 Flash: Whole-record check: current publisher specifications, licence, access or tariff (accessed 2026-10-25)
- DeepSeek V4 Flash: Whole-record check: current publisher specifications, licence, access or tariff (accessed 2026-10-25)
- DeepSeek V4 Flash: Whole-record check: current publisher specifications, licence, access or tariff (accessed 2026-10-25)
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 Claude Opus 5
- Claude Fable 5 vs Claude Opus 5
- Claude Haiku 4.5 vs Claude Opus 5
Diving deeper on one model? Claude Opus 5 · DeepSeek V4 Flash
Common questions
Claude Opus 5 vs DeepSeek V4 Flash
Answered from the verified figures on this page rather than general guidance.
Which has the larger context window, Claude Opus 5 or DeepSeek V4 Flash?
Do Claude Opus 5 and DeepSeek V4 Flash support the same reasoning levels?
Should I use Claude Opus 5 or DeepSeek V4 Flash?
Can I self-host Claude Opus 5 or DeepSeek V4 Flash?
Next step
Choosing between them
The verified figures that separate this pair, computed from the catalog rather than restated boilerplate.
- Deployment: DeepSeek V4 Flash publishes weights you can self-host; the other is API-only.
- Positioning: Claude Opus 5 sits in the frontier tier, DeepSeek V4 Flash in the balanced 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.