GLM-5V-Turbo vs Kimi K2.5

Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.

Agentic
Coding
Multimodal & Grounded
Reasoning
Knowledge
Instruction Following
Multilingual
Mathematics

GLM-5V-Turbo· Kimi K2.5

Quick Verdict

Pick Kimi K2.5 if you want the stronger benchmark profile. GLM-5V-Turbo only becomes the better choice if agentic is the priority or you need the larger 200K context window.

Kimi K2.5 is clearly ahead on the aggregate, 71 to 58. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-5V-Turbo is also the more expensive model on tokens at $1.20 input / $4.00 output per 1M tokens, versus $0.50 input / $2.80 output per 1M tokens for Kimi K2.5. GLM-5V-Turbo gives you the larger context window at 200K, compared with 128K for Kimi K2.5.

Operational tradeoffs

Price$1.20 / $4.00$0.50 / $2.80
SpeedN/A45 t/s
TTFTN/A2.38s
Context200K128K

Decision framing

BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.

Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.

BenchmarkGLM-5V-TurboKimi K2.5
AgenticGLM-5V-Turbo wins
BrowseComp51.9%60.6%
OSWorld-Verified62.3%63.3%
BrowseComp-VL51.9%42.9%
OSWorld62.3%63.3%
AndroidWorld75.7%43.1%
WebVoyager88.5%84.3%
Terminal-Bench 2.050.8%
Tau2-Airline80.0%
Tau2-Telecom95.9%
PinchBench84.8%
BFCL v468.3%
Coding
HumanEval99%
SWE-bench Verified76.8%
SWE-bench Verified*70.8%
LiveCodeBench85%
SWE-bench Pro40%
SWE-Rebench58.5%
React Native Evals74.9%
Multimodal & Grounded
Design2Code94.8%91.3%
Flame-VLM-Code93.8%88.8%
Vision2Web31.0%33.2%
ImageMining30.7%24.4%
MMSearch72.9%58.7%
MMSearch-Plus30.0%25.6%
SimpleVQA78.2%71.5%
Facts-VLM58.6%57.8%
V*89.0%84.3%
MMMU-Pro78.5%
OfficeQA Pro69%
Reasoning
MuSR72%
BBH81%
LongBench v267%
MRCRv270%
Knowledge
MMLU77%
GPQA87.6%
GPQA-D86.9%
SuperGPQA74%
MMLU-Pro87.1%
MMLU-Pro (Arcee)87.1%
HLE11%
FrontierScience67%
SimpleQA74%
Instruction Following
IFEval94%
IFBench70.2%
Multilingual
MGSM83%
MMLU-ProX78%
Mathematics
AIME 202377%
AIME 202479%
AIME 202578%
AIME25 (Arcee)96.3%
HMMT Feb 202373%
HMMT Feb 202475%
HMMT Feb 202574%
BRUMO 202576%
MATH-50082%
Frequently Asked Questions (2)

Which is better, GLM-5V-Turbo or Kimi K2.5?

Kimi K2.5 is ahead overall, 71 to 58. The biggest single separator in this matchup is AndroidWorld, where the scores are 75.7% and 43.1%.

Which is better for agentic tasks, GLM-5V-Turbo or Kimi K2.5?

GLM-5V-Turbo has the edge for agentic tasks in this comparison, averaging 58 versus 57.6. Inside this category, AndroidWorld is the benchmark that creates the most daylight between them.

Last updated: April 1, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.