Model comparison

DeepSeek-V3 vs Grok 4.5

Grok 4.5 is the stronger model overall, scoring 55.0 to 39.5 on the Noometry Index. DeepSeek-V3 costs 7.4× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.

Last verified October 10, 2026. 28 shared benchmarks.

DeepSeek-V3

DeepSeek

39.5

Rank #166 Confirmed

Grok 4.5 xAI

55.0

Rank #25 Confirmed

Summary

They share 28 benchmarks with published results for both. DeepSeek-V3 scores higher in 0 categories and Grok 4.5 in 8 categories; 8 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 20.5.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for DeepSeek-V3 and 97.8% for Grok 4.5.
DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $2 / $6 for Grok 4.5.
Grok 4.5 accepts more context: 500K tokens versus 164K.
DeepSeek-V3 has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3 and Grok 4.5 specifications
	DeepSeek-V3	Grok 4.5
Provider	DeepSeek	xAI
Noometry Index	39.5	55.0
Released	2024-12-26	2026-07-08
Weights	Open	Proprietary
Context window	164K	500K
Max output	164K	500K
Input $ / M tokens	$0.24	$2
Output $ / M tokens	$0.90	$6
Results tracked	60	52

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Grok 4.5 leads

DeepSeek-V3: 42.3 (#106), Grok 4.5: 52.2 (#35)

Coding benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
SciCode	35.8%	54.1%
WeirdML	36.1%	46.4%
LMArena Coding	1368	1474
DeepSWE	—	53.8%
FrontierCode	—	42.4%
Aider Polyglot	55.1%	—
LMArena WebDev	—	1553
BigCodeBench Instruct	50%	—
LiveBench Coding	70.9%	—
BigCodeBench Complete	62.2%	—
ALE-Bench	—	1,309
HumanEval+	86.6%	—
MBPP+	73%	—

Agentic & Tool Use Not comparable

DeepSeek-V3: —, Grok 4.5: 44.4 (#17)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
APEX-Agents	—	56.2%
τ²-bench Banking	—	47.9%
PostTrainBench	—	23.4%
GBAEval	—	65.4%
GDP.pdf	—	14%
LMArena Search	—	1213
METR Time Horizons	49.6%	—
Vending-Bench 2	—	3,887

Reasoning Grok 4.5 leads

DeepSeek-V3: 20.5 (#236), Grok 4.5: 56.1 (#25)

Reasoning benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
SimpleBench	27.2%	70%
Kagi LLM Benchmark	52.3%	83.5%
CritPt	0%	15.4%
LMArena Hard Prompts	1365	1462
DTBench	64.8%	96.5%
LMCA	15.5%	45.2%
Epoch Capabilities Index	135.94	153.92
ARC-AGI-2	—	52.6%
NYT Connections (extended)	—	79.9%
ARC-AGI-1	—	87.2%
Chess Puzzles	—	36%
LiveBench Reasoning	65.8%	—
LiveBench Data Analysis	60.9%	—
Surface Evolver Bench	—	74.4%
BIG-Bench Hard	87.5%	—
ForecastBench	59.1	—
HellaSwag	88.9%	—
LiveBench	66.9%	—
PIQA	84.7%	—
WinoGrande	85.2%	—

Math Grok 4.5 leads

DeepSeek-V3: 32.1 (#219), Grok 4.5: 60.9 (#35)

Math benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
OTIS Mock AIME 2024-2025	37.8%	97.8%
LMArena Math	1373	1459
FrontierMath (Tiers 1-3)	—	57.2%
FrontierMath Tier 4	—	24.4%
ProofBench	—	31%
Omni-MATH	40.3%	—
LiveBench Math	73.5%	—
MATH Level 5	75.5%	—
FrontierMath (Feb 2025 set)	1.7%	—

Knowledge Grok 4.5 leads

DeepSeek-V3: 37.5 (#155), Grok 4.5: 62.3 (#24)

Knowledge benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
GPQA Diamond	67.6%	93.4%
LMArena Expert	1351	1466
SimpleQA Verified	—	48.3%
MMLU-Pro	72.3%	—
Confabulations	26.1%	—
Vectara Hallucination Rate	6.1%	—
GPQA (HELM)	53.8%	—
ARC (AI2) Challenge	95.3%	—
MMLU	87.2%	—
TriviaQA	82.9%	—

Multimodal Not comparable

DeepSeek-V3: —, Grok 4.5: 37.6 (#72)

Multimodal benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
LMArena Vision	—	1288
Blueprint-Bench 2	—	27.3%
Furniture Assembly	—	22.5%
LMArena Document	—	1452

Multilingual Grok 4.5 leads

DeepSeek-V3: 48.5 (#143), Grok 4.5: 54.4 (#42)

Multilingual benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
LMArena Non-English	1358	1440
LMArena Chinese	1391	1496
LMArena French	1385	1456
LMArena German	1374	1446
LMArena Japanese	1333	1428
LMArena Korean	1319	1404
LMArena Russian	1373	1448
LMArena Spanish	1358	1450

Instruction Following Grok 4.5 leads

DeepSeek-V3: 72.8 (#130), Grok 4.5: 76.0 (#48)

Instruction Following benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
LMArena Instruction Following	1345	1446
LiveBench Instruction Following	81.5%	—
IFEval	83.2%	—

Long Context Grok 4.5 leads

DeepSeek-V3: 34.0 (#253), Grok 4.5: 44.8 (#56)

Long Context benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
LMArena Longer Query	1352	1463
Fiction.LiveBench	50%	—

Writing & Preference Grok 4.5 leads

DeepSeek-V3: 57.4 (#130), Grok 4.5: 65.8 (#42)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3	Grok 4.5
LMArena Text	1375	1448
LMArena Creative Writing	1364	1442
EQ-Bench Creative Writing	1472	1579
LMArena Multi-Turn	1389	1456
Short-Story Creative Writing	77%	—
WildBench	83%	—
LiveBench Language	49.1%	—

Frequently asked questions

Is DeepSeek-V3 better than Grok 4.5?

Which is cheaper, DeepSeek-V3 or Grok 4.5?

DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Grok 4.5 lists at $2 and $6.

Is DeepSeek-V3 or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 42.3 in the Noometry coding category.

Which has the bigger context window?

Grok 4.5 does, with 500K tokens against 164K.

How many benchmarks do DeepSeek-V3 and Grok 4.5 share?

28 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Grok 4.5 has 52.