DeepSeek V4 Pro: score and benchmark evidence
Historical release · · DeepSeek
83.2 points · Rank 15 among 19 ranked models · Stability range 74.1–93.9 points.
Rank follows the point estimate; it does not establish statistically significant superiority. The scale is anchored at mean 100, SD 15 in the frozen calibration cohort, not human IQ or a percentage.
One capability score combines evidence across seven equally weighted domains. Related benchmark variants share a family budget, and influence from one evaluator is capped.
Release tli-2026-v1.0-2026-09-06. Methodology and limitations for this release.
This page preserves historical evidence. See the current model record.
Capability domains
Domain scores use calibration-cohort standard-deviation units. Unmeasured domains are shown as unavailable, not zero.
| Domain | Score (SD units) | Observed benchmarks |
|---|---|---|
| Knowledge & reasoning | -0.50 | 10 |
| Coding & software engineering | -0.18 | 8 |
| Agentic tool & computer use | -0.62 | 3 |
| Professional & real-world work | -0.68 | 13 |
| Multimodal & vision | Unavailable | 0 |
| Cybersecurity | Unavailable | 0 |
| Preference & communication | -0.47 | 1 |
Published benchmark evidence
35 results contribute to this score, from 56 collected results and 34 contributing benchmark families. Counts are not independent sample sizes.
Collected means a numeric result exists in the database. Used means its source, exact model checkpoint and benchmark protocol have been reviewed and its benchmark has a frozen calibration supported by multiple providers. For each benchmark we select the highest published aggregate across comparable reasoning modes, not necessarily maximum effort, and count it once. More tested settings provide more chances to record a high score; this selection advantage is not corrected by the conditional interval. Developer-reported and sole/unspecified settings remain eligible and disclosed. Unresolved does not mean false or useless: conflicting metrics, unverified citations and unmatched versions stay collected until reconciled.
The bar estimates score stability under the scoring assumptions: a fixed calibration and comparable benchmark availability. It resamples benchmark families and evaluators while keeping each domain’s evidence strength fixed, then adds a frozen correction learned from missing-evidence tests. Benchmark count and domain coverage both affect that correction. It does not add a second full capability-posterior draw or uncertainty from changing the reference calibration. It is not a guarantee under selective publication, new benchmark domains or future model releases.
Reporting sensitivity: ±1.4 points. Sensitivity to selectively published results. This stress range is not a confidence interval or a probability statement.
GDPval-AA v2 (Elo): 1496.68 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: Elo
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
Terminal-Bench 2.1: 78.651685%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: pass@1, AA harness
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
DeepSWE v1.1: 62.831858%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek-v4-pro; max; mini-swe-agent
Metric: pass@1 over scored rollout attempts
Original source · Retrieved 2026-09-06
Selected the best pass@1 across published reasoning efforts within mini-swe-agent. Pass@4 is not substituted.
LMArena (Chatbot Arena) — Text: 1459.610748 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek-v4-pro-high-20260813; Arena text overall; style control enabled
Metric: style-controlled text overall Elo
Original source · Retrieved 2026-09-06
Best published reasoning-mode rating within the same text overall board. Different leaderboard slices and undated earlier DeepSeek checkpoints are not pooled.
AA Intelligence Index v4.1.1 (score): 53 native-points
Not used in the score. Source category not recorded. Review status: excluded.
Held-out composite, never a scoring input
GPQA Diamond: 92.828283%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: accuracy, AA
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
AA-LCR: 75.3%
Not used in the score. Source category: independent. Review status: unresolved.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Original source · Retrieved 2026-09-06
Current AA result is LCR v1.1. The inherited AA-LCR row has no reviewed version match; a new version cannot silently substantiate the older value.
Finance Agent v2 — Vals: 50.39%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: fabv2 overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Excel Modeling Benchmark — Vals: 52.8%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: emb overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
SciCode: 51.041667%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: accuracy, AA harness
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
Humanity's Last Exam (no tools): 41.010195%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: accuracy, no tools, AA
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
Legal Research Bench — Vals: 40.87%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: legal_research overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Harvey LAB (Vals): 7.5%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: hlab overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
AA-Briefcase Overall Elo: 1269.34 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: overall Elo
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
MedScribe — Vals: 80.17%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: medscribe overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
SWE-bench Verified — Vals: 96.4%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: swebench overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
LegalBench — Vals: 82.36%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: legal_bench overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
TaxEval v2 — Vals: 73.06%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: tax_eval_v2 overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
KingBench 3: 76.25%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 76.25 for DeepSeek V4 Pro Max on KingBench 3 still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
Vals Index: 66.25%
Not used in the score. Source category: benchmark_org. Review status: excluded.
Configuration: deepseek-v4-pro; max effort; Vals AI index methodology
Metric: Composite score
Original source · Retrieved 2026-08-16T00:00:00.000Z
Held-out composite, never a scoring input
Public Benefits Bench — Vals: 62.92%
Contributes to the score. Source category: benchmark_org. Review status: reviewed-archive.
Configuration: deepseek-v4-pro; max effort; Vals AI public-benefits evaluation
Metric: Accuracy
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
MedCode — Vals: 42.47%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: medcode overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
AutomationBench v1.0.6: 43.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 43.2 for DeepSeek V4 Pro Max on AutomationBench v1.0.6 still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
AA-Omniscience: 0.833333 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: knowledge reliability index -100 to 100
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
MMLU Pro — Vals: 86.97%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: mmlu_pro overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
LiveBench: 77.436143%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek-v4-pro-0813; LiveBench 2026-06-25 release
Metric: official overall: equal mean of seven category means
Original source · Retrieved 2026-09-06
Task mapping archived from categories_2026_06_25.json. Aggregate follows the official UI, not a flat mean of the 23 tasks. Undated old DeepSeek checkpoints excluded.
CorpFin v2 — Vals: 65.42%
Contributes to the score. Source category: benchmark_org. Review status: reviewed-archive.
Configuration: deepseek-v4-pro; max effort; Vals AI CorpFin v2 evaluation with Sonnet 4.5 judge
Metric: Accuracy
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
Code Migration †: 41.54%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: code-migration overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Toolathlon-Verified: 74.4%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (max) ✓; Default agent
Metric: pass@1, verified tasks
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Agents' Last Exam: 12.4%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro Max; High; OpenClaw
Metric: pass rate, ALE-V1 147 tasks; not partial score
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
LiveCodeBench: 87.53%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: lcb overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
CyberGym: 83.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 83.3 for DeepSeek V4 Pro Max on CyberGym still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
Humanity's Last Exam w/ tools: 60%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 60 for DeepSeek V4 Pro Max on Humanity's Last Exam w/ tools still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
SWE-Marathon v1.1: 10.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 10.6 for DeepSeek V4 Pro Max on SWE-Marathon v1.1 still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
ProgramBench v1 (Raw Pass Rate) — Vals: 70.103%
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: deepseek/deepseek-v4-pro-0813 · reasoning_effort=max
Metric: Raw Pass Rate: mean percent of hidden behavioral tests passed per task. Not Fully Resolved (all tests pass) and not Almost Resolved (≥95% of tests pass).
Original source · Retrieved 2026-09-05
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
FrontierMath v2 Tier 4: 26.83%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 26.83 for DeepSeek V4 Pro Max on FrontierMath v2 Tier 4 still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
SWE-bench Pro: 55.4%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 55.4 for DeepSeek V4 Pro Max on SWE-bench Pro still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
BrowseComp: 83.4%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 83.4 for DeepSeek V4 Pro Max on BrowseComp still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
ARC-AGI-2: 61.3%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro Max; Max; ARC Prize standard harness
Metric: semi-private verified score percentage
Original source · Retrieved 2026-09-06
Best reasoning variant within the verified standard-harness summary. Provider Adapter, public tasks and community agent scores are not substituted.
SimpleQA: 57.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 57.9 for DeepSeek V4 Pro Max on SimpleQA still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
MCP Atlas †: 73.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 73.6 for DeepSeek V4 Pro Max on MCP Atlas † still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
NL2Repo: 61.1%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 61.1 for DeepSeek V4 Pro Max on NL2Repo still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
CoWorkBench: 66.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 66.3 for DeepSeek V4 Pro Max on CoWorkBench still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
τ²-Bench Telecom: 96.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 96.2 for DeepSeek V4 Pro Max on τ²-Bench Telecom still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
FrontierCode v1.1 Main — Anthropic H2H: 17.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 17.6 for DeepSeek V4 Pro Max on FrontierCode v1.1 Main — Anthropic H2H still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
EQ-Bench 3: 1570.2 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 1570.2 for DeepSeek V4 Pro Max on EQ-Bench 3 still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
τ³-Banking: 39.587629%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: strict task success, AA
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
CritPt: 18%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: accuracy, AA official grading
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
ProofBench v1.1 — Vals: 50%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: proof_bench overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Vibe Code Bench v1.1 — Vals: 82.3%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: vibe-code overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
ARC-AGI-1: 90.5%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro Max; Low; ARC Prize standard harness
Metric: semi-private verified score percentage
Original source · Retrieved 2026-09-06
Best reasoning variant within the verified standard-harness summary. Provider Adapter, public tasks and community agent scores are not substituted.
EnterpriseOps-Gym-AA: 49.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 49.6 for DeepSeek V4 Pro Max on EnterpriseOps-Gym-AA still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
MLCR-AA (Medical Long Context Reasoning): 17.777778%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: DeepSeek V4 Pro 0813 (Reasoning, Max Effort); Artificial Analysis published evaluation harness
Metric: overall accuracy, AA
Original source · Retrieved 2026-09-06
Best published result among 1 reasoning configurations of the exact AA model release; fallback configurations excluded.
IFBench: 76.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 76.5 for DeepSeek V4 Pro Max on IFBench still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
Mystery Game Puzzles v1.0.4 — Epoch: 43%
Not used in the score. Source category not recorded. Review status: unresolved.
Not admitted after the September 6 model review: the collected 43 for DeepSeek V4 Pro Max on Mystery Game Puzzles v1.0.4 — Epoch still lacks a matched source entry establishing its checkpoint and benchmark-specific metric, tools and retry setup. This is a citation/comparability gap, not a finding that the result is false or useless. More or maximum reasoning effort is not required.
Tax Agent Bench v1 — Vals: 58.66%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: deepseek_deepseek-v4-pro-0813; Vals max effort
Metric: tax_agent_bench overall percentage in Vals benchmark harness
Original source · Retrieved 2026-09-06
Reconciled against the exact model/configuration in the evaluator’s public source. Updated source value recorded explicitly; best published reasoning mode selected within the same benchmark protocol.
Cite this model record
The Latent. “DeepSeek V4 Pro: score and benchmark evidence.” 2026-09-06. Release tli-2026-v1.0-2026-09-06.
Download release data and definitions (JSON) · Data usage terms
Related reporting
These dated stories provide reporting context; they may describe an earlier release.