DeepSeek V4.1 Flash
DeepSeek · Released Sep 2026
DeepSeek V4.1 Flash, the model now behind DeepSeek's deepseek-flash API, is a strong-value open-weight pick: MIT-licensed weights, image input, a 1M-token context and an Artificial Analysis score well above average for comparable models. Conditional because its standout coding-agent results are DeepSeek's own, Artificial Analysis measured it as very verbose, and peak hours double the rate. Right for cost-sensitive pipelines that can run off-peak or self-host; wrong for buyers who need independently verified agent performance before committing.
Is it right for you?
Good for
- High-volume API workloads at $0.15 input and $0.60 output per 1M tokens off-peak, with off-peak cache-hit input at $0.003 per 1M
- Scheduling around peak pricing: peak rates, twice off-peak, apply only 01:00-04:00 and 06:00-10:00 UTC Monday through Friday, and all other hours bill off-peak
- Self-hosting under the MIT licence, with the weights published on Hugging Face
- Terminal-style coding agents: DeepSeek's own max-effort table reports 90.6 on Terminal-Bench 2.1 and 74.2 on DeepSWE v1.1, the top score on both rows, ahead of Opus-5.0's 89.1 and 74.0
- Document and image understanding: Artificial Analysis lists text and image input, and DeepSeek's base-model table reports 95.6 on DocVQA
- Long-context work within a 1M-token window and up to 384K output tokens
Not good for
- Terminal-Bench 3.0 and 4.0: DeepSeek's own table reports 30.0 and 31.2, against Opus-5.0's 43.3 and 51.8
- Hard expert-knowledge questions: DeepSeek's table reports 36.8 on HLE, against Opus-5.0's 56.3 and GPT-5.6 Sol's 44.5
- Tight output-token budgets: Artificial Analysis measured 250M output tokens to run its Intelligence Index, against a 140M median for open-weight models of similar size
- Assuming the headline agent score carries over to your harness: DeepSeek's own scaffold table ranges from 65.5 under OpenCode to 74.2 under mini-SWE on DeepSWE v1.1
- Broad factual recall: DeepSeek's base-model table reports 42.3 on SimpleQA-Verified, against 55.2 for V4-Pro-Base
- Image generation: Artificial Analysis lists text as its only output modality
- Treating the stored benchmark figures as current: DeepSeek's pricing page now lists the served version as DeepSeek-V4.1-Flash-1017, and its October 17, 2026 changelog says that update has stronger benchmark results and shorter outputs. The scores here are from the September 10 release, and we have not read the 1017 scores
How it performs by task
Terminal-style agentic coding
Tops DeepSeek's own max-effort table on Terminal-Bench 2.1 (90.6) and DeepSWE v1.1 (74.2). Vendor-run figures.
Terminal-Bench 3.0 and 4.0
31.2 on Terminal-Bench 4.0 in DeepSeek's table, against 51.8 for Opus-5.0.
Cost-efficient high-volume inference
$0.15/$0.60 per 1M off-peak, but Artificial Analysis records 250M output tokens on its index run against a 140M similar-size median.
Visual document understanding
95.6 on DocVQA in DeepSeek's base-model table; image input, text output. No independent vision measurement found.
Expert-level reasoning (HLE)
36.8 on HLE, against Opus-5.0's 56.3 in DeepSeek's table.
Pricing
Input
$0.15 / 1M tokens
Output
$0.60 / 1M tokens
Context
1M tokens; 2x price at weekday peak hours
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| Artificial Analysis Intelligence Index | 40 | Source |
| Terminal-Bench 2.1 (DeepSeek-run) | 90.6 | Source |
| Terminal-Bench 4.0 (DeepSeek-run) | 31.2 | Source |
| DeepSWE v1.1 (DeepSeek-run) | 74.2 | Source |
| GPQA Diamond (DeepSeek-run) | 90.9 | Source |
| HLE (DeepSeek-run) | 36.8 | Source |
| CyberGym (DeepSeek-run) | 88.1 | Source |
| Codeforces rating (DeepSeek-run) | 3471 | Source |
No verdict changes yet
The clock starts day one. Changes land here as our verdict evolves.
Sources
- Artificial Analysis, DeepSeek V4.1 Flash (Reasoning, Max Effort)Sep 2026
- DeepSeek API Docs: DeepSeek-V4.1-Flash Release (2026/09/10)Oct 2026
- Hugging Face, deepseek-ai/DeepSeek-V4.1-Flash model card (official, MIT)Sep 2026
- DeepSeek API Docs: Models and PricingOct 2026
- DeepSeek API changelog (V4 Pro API service continues after Sep 14, 2026)Oct 2026
- DeepSeek API Docs: DeepSeek-V4.1-Flash-1017 update (2026/10/17)Oct 2026
Verification log
- Profile· Updated
Automated agent
DeepSeek shipped DeepSeek-V4.1-Flash-1017 on 2026-10-17 (stronger benchmarks, shorter outputs per changelog); pricing page lists it as the served version. Price unchanged. Added an avoid_for caution that stored benchmarks are from the Sep 10 release, added changelog sources. Draft post deepseek-v4-1-flash-1017-update-v4-pro-stays-on-api.
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· Updated
Automated agent
First log row for this entity. Rates unchanged: deepseek-flash = DeepSeek-V4.1-Flash, 1M ctx, 384K max output, $0.15/$0.60 off-peak, $0.3/$1.2 peak, cache hit $0.003/$0.006. REFRESHED pricing_context '1M ctx. Peak 2x: 01-04+06-10 UTC Mon-Fri' -> '1M ctx. Peak 2x 01-04+06-10 UTC M-F, ex CN hols' (footnote (2) now excludes Chinese public holidays; two reads). First attempt at 57 chars failed with a raw asyncpg StringDataRightTruncationError (pricing_context is varchar(50), unvalidated at the tool boundary). entity_id taken from that error's SQL parameters, since no read tool returns it (gap f88981cc). Strength row 'peak rates ... apply only ... Monday through Friday' left as is (still accurate on non-holiday weekdays). Stale shared source label 'Models & Pricing (V4 Pro: $0.44/$0.87)' relabelled by this record (7ef6c662).