Models · DeepSeek · Out sinceReleased 21 Aug 2026
DeepSeek V4 Flash Vision Exp
DeepSeek V4 Flash Vision Exp is made by DeepSeek. We don't have enough test results yet to rank it. It's cheap to use. Its makers have published it, so you can run it on your own servers.
Experimental vision-enabled V4-Flash. API name now routes to deepseek-flash (V4.1-Flash) per pricing page.
- —
- —
- —
- Cheap$0.22 / $0.65
- $0.32
- 1M
- —
- 19 (12 independent12 indep.)
- 21 Aug 2026
- DeepSeek
- text, image
Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.
OpenRouter is a service that gives access to many AI models in one place. This model is listed there as deepseek/deepseek-v4-flash-vision-exp.
Route it as deepseek/deepseek-v4-flash-vision-exp at $0.22 in / $0.65 out per 1M tokens, 1M context. Listed since 21 Aug 2026.
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Run it yourself
Self-hosting
On your own servers,Licence, sizeunder your own control.and hardware.
Details comingselfHost data pending
Every test result
Every result
All the numbers,with where they came from.
Every result we've found for this model, with who measured it and a link to where we read it.
All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.
| NotesNotes | ||||||
|---|---|---|---|---|---|---|
| AA-LCR | 81.3% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-LCR accuracy, AA-run (long multi-document reasoning) |
| AA-Omniscience Accuracy | 38.6% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | |
| AA-Omniscience Hallucination Rate | 91.5% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page) |
| AA-Omniscience Index | -17.6 | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised |
| Agents' Last Exam | 27.3% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | — | temperature=1.0, top_p=0.95 |
| Artificial Analysis Intelligence Index | 34.8 | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA slug deepseek-v4-flash-vision; list price $0.44/1.32 per 1M in/out; cost to run AA Intelligence Index $0.31/task |
| Artificial Analysis output speed | 211 tok/s | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | tokens/sec (median output speed, first-party API); TTFT 0.9s; list price $0.44/1.32 per 1M in/out |
| AutomationBench | 25.7% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | — | temperature=1.0, top_p=0.95 |
| CyberGym | 75.3% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | DeepSeek Harness (minimal mode) | temperature=1.0, top_p=0.95 |
| DeepSWE | 59.3% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | DeepSeek Harness (minimal mode) | temperature=1.0, top_p=0.95 |
| FrontierSWE | 14.8% | Default | Independent testIndependentFrontierSWE ↗ | 1 Oct 2026 | proximus | FrontierSWE V2, mean@5 over 34 tasks (20h budget); ±9.7; $8.57/trial; 14.8h/trial; 'Per provider' view shows best entry per provider; Epoch notes runs use max reasoning effort |
| GPQA Diamond | 91.3% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-run evaluation (AA slug deepseek-v4-flash-vision) |
| Humanity's Last Exam | 34.5% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-run evaluation (AA slug deepseek-v4-flash-vision) |
| NL2Repo-Bench | 57.7% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | DeepSeek Harness (minimal mode) | temperature=1.0, top_p=0.95 |
| SciCode | 49.7% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-run evaluation (AA slug deepseek-v4-flash-vision) |
| Terminal-Bench 2.1 | 74.2% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-run evaluation (AA slug deepseek-v4-flash-vision) |
| Terminal-Bench 2.1 | 83.9% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | DeepSeek Harness (minimal mode) | temperature=1.0, top_p=0.95 |
| Terminal-Bench 4.0 | 12.1% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | 1 Oct 2026 | — | AA-run evaluation (AA slug deepseek-v4-flash-vision) |
| Toolathlon-Verified | 75.9% | Maxreasoning_effort=max | Maker's own figureVendor-reportedDeepSeek ↗ | 21 Aug 2026 | — | temperature=1.0, top_p=0.95 |