GenMagic logoGenMagic
Sign in

DeepSeek V4 Pro 0423 vs GPT-5.6 Terra

DeepSeek against OpenAI. Prices, scores and speed below are live.

  • DeepSeek V4 Pro 0423 costs 56% less on blended price ($3.00 against $6.75 per 1M tokens, 3 input to 1 output).
  • GPT-5.6 Terra scores higher on the Artificial Analysis Intelligence Index: 32.2 against 30.9.
  • On the 26 benchmarks both have scores on, DeepSeek V4 Pro 0423 leads on 7 and GPT-5.6 Terra on 19.
  • GPT-5.6 Terra reads more context: 1.1M tokens against 1M.
  • GPT-5.6 Terra is newer: listed Jul 9, 2026, against Apr 24, 2026.
Run both on your promptDeepSeek V4 Pro 0423GPT-5.6 Terra

Side by side

DeepSeek V4 Pro 0423GPT-5.6 Terra
LabDeepSeekOpenAI
ListedApr 24, 2026Jul 9, 2026
Input per 1M tokens$2.40$3.00
Output per 1M tokens$4.80$18.00
Cached input per 1M$0.20$0.30
Blended (3:1) per 1M$3.00$6.75
Context window1M1.1M
Max output393K128K
Knowledge cutoff-2026-02-16
Intelligence index30.932.2
Coding index59.476.7
ReasoningYesYes
Reads imagesNoYes
Tool useYesYes
Structured outputYesYes

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkDeepSeek V4 Pro 0423GPT-5.6 Terra
AIME46.7%Epoch AI53.3%Epoch AI
Artificial Analysis Agentic Index26.3Artificial Analysis and OpenRouter43.2Artificial Analysis and OpenRouter
Artificial Analysis Coding Index59.4Artificial Analysis and OpenRouter76.7Artificial Analysis and OpenRouter
Artificial Analysis Intelligence Index30.9Artificial Analysis and OpenRouter42.3Artificial Analysis and OpenRouter
Chess Puzzles19%Epoch AI47%Epoch AI
DTBench90.7%Epoch AI93.3%Epoch AI
FrontierCode17.6Epoch AI41.3Epoch AI
FrontierMath Tier 4 (v2)2.4%Epoch AI70.7%Epoch AI
FrontierMath Tiers 1-3 (v2)45.3%Epoch AI86%Epoch AI
GPQA Diamond87.1%Artificial Analysis and OpenRouter88.4%Artificial Analysis and OpenRouter
LMArena (Coding)1469 EloLMArena1485 EloLMArena
LMArena (Overall)1451 EloLMArena1448 EloLMArena
LMArena: Creative Writing1446 EloLMArena1413 EloLMArena
LMArena: Hard Prompts1461 EloLMArena1469 EloLMArena
LMArena: Longer Query1457 EloLMArena1453 EloLMArena
LMArena: Math1455 EloLMArena1477 EloLMArena
LMArena: Multi-Turn1467 EloLMArena1455 EloLMArena
LMCA41.2Epoch AI55Epoch AI
Mystery Game Puzzles17%Epoch AI10%Epoch AI
ProofBench16%Epoch AI74%Epoch AI
SciCode50%Epoch AI55%Epoch AI
SimpleQA Verified47%Epoch AI43.2%Epoch AI
Surface Evolver Bench40Epoch AI83.8Epoch AI
tau-bench (Airline)76.8%Artificial Analysis and OpenRouter72.3%Artificial Analysis and OpenRouter
WebDev Arena1464 EloEpoch AI1521 EloEpoch AI
WeirdML48.9%Epoch AI78.3%Epoch AI

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See DeepSeek models, OpenAI models, every LLM's price and our sources.