DeepSeek V4 Flash 0423

DeepSeek V4 Flash 0423 is a model from DeepSeek. 17 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score. 50 is the middle.63.0 ±5.4
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.24/s
Input / 1MUS dollars per 1M input tokens.$0.089
Output / 1MUS dollars per 1M output tokens.$0.177
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.1432 (#83)

50 is the middle of the board. The range shows the doubt in the index.

48,887 votes. Elo shows what people prefer. It does not change the score.

CapabilitiesScore per category. 50 is the middle.

50 is the middle
AgenticMulti-step tasks with tools.
62.0
CodingCode writing and repair.
65.5
ReasoningLogic problems and puzzles.
61.7
MultimodalTasks with images and text.
N/A
KnowledgeFacts and expert knowledge.
59.4
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
61.6
MathMath problems.
65.6

Results

17 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result on the index scale.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
GPQA DiamondKnowledge89.9%61.21 Sept 2026Vals AI
SWE-benchCoding88.8%68.81 Sept 2026Vals AI
LiveCodeBenchCoding87.3%63.61 Sept 2026Vals AI
MMLU ProKnowledge86.2%56.41 Sept 2026Vals AI
LiveBench MathematicsMath83.2%59.125 Jun 2026LiveBench
LiveBench ReasoningReasoning78.6%65.225 Jun 2026LiveBench
Vibe Code Bench v1.1Coding74.7%73.3OpenHands21 Sept 2026Vals AI
LiveBench LanguageKnowledge74.6%60.725 Jun 2026LiveBench
LiveBench Data AnalysisReasoning73.7%58.325 Jun 2026LiveBench
LiveBench CodingCoding72.1%57.425 Jun 2026LiveBench
Terminal-Bench 2.1Agentic67.0%63.621 Sept 2026Vals AI
LiveBench Instruction FollowingInstruction64.3%61.625 Jun 2026LiveBench
ProofBench v1.1Math56.0%72.121 Sept 2026Vals AI
SkillsBenchCoding50.7%65.9OpenHands11 Sept 2026Vals AI
LiveBench Agentic CodingAgentic42.2%56.425 Jun 2026LiveBench
Code MigrationCoding38.6%69.621 Sept 2026Vals AI
IOICoding32.7%59.921 Sept 2026Vals AI
Terminal-Bench 4.0Agentic9.1%65.921 Sept 2026Vals AI
ProgramBenchCoding0.0%21 Sept 2026Vals AI

17 benchmarks count, from 18 of 19 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authorsOpenRouter, collected directlyNo licence statedVals AI, collected directlyNo licence stated. Read from the public leaderboard and credited to Vals AILiveBench, collected directlyNo licence stated for the leaderboard. The site repo serving these CSVs has no LICENSE; the harness repo carries upstream Apache-2.0 and MIT copies that cover code, not results

Same level, lower price

Qwen3.8-Flash-Next66.4 · Freedots3-note Preview65.5 · Free

More from DeepSeek

DeepSeek V4.1 Flash70.6DeepSeek V4 Pro 081368.8DeepSeek V4 Flash 073164.6DeepSeek V4 Pro 042362.4DeepSeek V3.2 (Thinking)49.9