DeepSeek-V4.1-Flash-1017
- Input price
- $0.15
- Per 1 million tokens (USD)
- Output price
- $0.60
- Per 1 million tokens (USD)
- Effort levels
- 3total
- Default: High
- Benchmark results
- 9total
- Benchmark types: 9
Model information
- API model ID
- deepseek-flash
- release date
- 2026-10-17
- effort
- Low / High (default) / Max
- Effort source
- Based on DeepSeek thinking mode documentation (api-docs.deepseek.com/guides/thinking_mode).
- Claude Code
- unavailable
- Codex
- unavailable
- Availability notes
- DeepSeek official documentation provides connection instructions for Claude Code (mapping claude-haiku/sonnet to deepseek-flash) and Codex. The connection instructions are in the model provider documentation, not official model options for Claude Code / Codex, so this is false.
- Availability source
- https://api-docs.deepseek.com/quick_start/agent_integrations/codex
Price (USD per 1 million tokens)
Prices do not vary by effort; token consumption does.
| input | $0.15 |
|---|---|
| output | $0.60 |
| cache read | $0.003 |
| cache write | — |
Off-peak pricing. Peak is 2x: cache hit $0.006 / miss $0.3 / output $1.2. The former name deepseek-v4-flash also redirects to the same model
deepseek-flash cache hit $0.003 / $0.006, cache miss $0.15 / $0.3, output $0.6 / $1.2 (Off-peak / Peak)Benchmark results
| Benchmark | effort | Score | Source cost (metric) | measured/reported by | source | Conditions and extraction method |
|---|---|---|---|---|---|---|
| Terminal-Bench 4.0not stated | not stated | 42.3% | — | DeepSeek | DeepSeek-V4.1-Flash-1017 (Hugging Face model card, deepseek-ai)official · Retrieved 2026-11-13 | Vendor-reported value (self-reported by DeepSeek). Card note: We use the Claude Code harness for Terminal-Bench 4.0 and the Mini-SWE-V2 harness for DeepSWE v1.1. Effort not stated. Claude Code harness.transcribed from tableSource excerptTerminal-Bench 4.0 | 42.3Location Comparison table at the start of the HF model card, Terminal-Bench 4.0 row, DS-V4.1-Flash-1017 column (leftmost numeric column) |
| DeepSWE v1.1not stated | not stated | 77.0% | — | DeepSeek | DeepSeek-V4.1-Flash-1017 (Hugging Face model card, deepseek-ai)official · Retrieved 2026-11-13 | Vendor-reported value (self-reported by DeepSeek). Card note: We use the Claude Code harness for Terminal-Bench 4.0 and the Mini-SWE-V2 harness for DeepSWE v1.1. Effort not stated. Mini-SWE-V2 harness.transcribed from tableSource excerptDeepSWE v1.1 | 77.0Location Comparison table at the start of the HF model card, DeepSWE v1.1 row, DS-V4.1-Flash-1017 column (leftmost numeric column) |
| GPQA DiamondMax | Max | 93.2% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptgpqa_diamond.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.9318181818181818, stderr=0.015671362011907143, Started at=2026-10-19T22:38:32.000ZLocation gpqa_diamond.csv, data row 12 (file line 24), id=ZK85PJtsCSeQJgV8dA4R4G, Model version=deepseek-v4.1-flash-1017_max |
| FrontierMath Tiers 1-3(v2)Max | Max | 71.2% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptfrontiermath_tiers_1_3_v2.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.712280701754386, stderr=0.02686277751301251, Started at=2026-10-19T22:39:00.000ZLocation frontiermath_tiers_1_3_v2.csv, data row 5 (file line 6), id=UnaMiD4Yy8W4WzyeSRwybr, Model version=deepseek-v4.1-flash-1017_max |
| FrontierMath Tier 4(v2)Max | Max | 46.3% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptfrontiermath_tier_4_v2.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.4634146341463415, stderr=0.07884502378168105, Started at=2026-10-19T22:39:00.000ZLocation frontiermath_tier_4_v2.csv, data row 5 (file line 6), id=jnCfi6VttQBv3t54B3xp9G, Model version=deepseek-v4.1-flash-1017_max |
| OTIS Mock AIME 2024-2025Max | Max | 99.4% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptotis_mock_aime_2024_2025.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.9944444444444445, stderr=0.003883473863441843, Started at=2026-10-19T22:39:00.000ZLocation otis_mock_aime_2024_2025.csv, data row 12 (file line 24), id=chayEvHvpt2Wcif7pg64hF, Model version=deepseek-v4.1-flash-1017_max |
| SimpleQA VerifiedMax | Max | 42.5% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptsimpleqa_verified.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.425, stderr=0.01564032031704017, Started at=2026-10-19T22:38:54.000ZLocation simpleqa_verified.csv, data row 12 (file line 24), id=Ei4XWaCH5drYUnDRo2MYCW, Model version=deepseek-v4.1-flash-1017_max |
| Chess Puzzles(Epoch AI)Max | Max | 46.0% | — | Epoch AI | Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0third-party · Retrieved 2026-11-12 | Run by Epoch AI (third-party measurement). Source label "DeepSeek V4.1 Flash 1017 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from tableSource excerptchess_puzzles.csv: Model version=deepseek-v4.1-flash-1017_max, Best score (across scorers)=0.46, stderr=0.05009082659620331, Started at=2026-10-19T22:39:00.000ZLocation chess_puzzles.csv, data row 11 (file line 23), id=4sRo7TxavQAsjaEALPZdDy, Model version=deepseek-v4.1-flash-1017_max |
| Epoch Capabilities Index(ECI)not stated | not stated | 159.41 | — | Epoch AI | Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)third-party · Retrieved 2026-11-12 | A per-model metric that combines results across multiple benchmarks and multiple efforts (not per effort). Calculated by Epoch AI.transcribed from tableSource excerpteci_scores.csv: Model=DeepSeek V4.1 Flash 1017, eci=159.41, eci_ci_low=157.01, eci_ci_high=162.0Location epoch_capabilities_index/eci_scores.csv, data row 22 (file line 23), Model=DeepSeek V4.1 Flash 1017 |
Related rankings
- Overall ranking (all models, records)
- Terminal-Bench 4.0 ranking by metric (all models, records)
- DeepSWE v1.1 ranking by metric (all models, records)
- GPQA Diamond ranking by metric (all models, records)
- FrontierMath Tiers 1-3(v2) ranking by metric (all models, records)
- FrontierMath Tier 4(v2) ranking by metric (all models, records)
- OTIS Mock AIME 2024-2025 ranking by metric (all models, records)
- SimpleQA Verified ranking by metric (all models, records)
- Chess Puzzles(Epoch AI) ranking by metric (all models, records)
- Epoch Capabilities Index(ECI) ranking by metric (all models, records)
- Recommendations by role