GLM-5 vs Kimi K3.1 Pro
Full profiles: GLM-5 (Zhipu AI) and Kimi K3.1 Pro (Moonshot AI): every benchmark, price and rank.
Updated · Data from 11 sources
As of 16 Nov 2026, GLM-5 and Kimi K3.1 Pro are neck and neck on quality (60.2 vs 61.8 on the BenchLeader Index, inside the ±4.7 margin), and GLM-5 is 2.6× cheaper at $1.55 vs $4.00 per 1M tokens blended. Choose GLM-5 for reasoning (68 vs 59) and fast first answers (29.2 s vs 157.4 s), where it also streams ~2× faster; choose Kimi K3.1 Pro for maths (73 vs 63), agents & tools (65 vs 58), and long context (68 vs 62).
Key facts (as of 16 Nov 2026)
- GLM-5 scores 60.2 on the BenchLeader Index and Kimi K3.1 Pro scores 61.8 (BenchLeader, 16 Nov 2026).
- GLM-5 costs $1.55 per 1M tokens blended (list price, 16 Nov 2026).
- Kimi K3.1 Pro costs $4.00 per 1M tokens blended (list price, 16 Nov 2026).
- GLM-5 streams ~2× faster than Kimi K3.1 Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
- Of the 5 benchmarks both models have, Kimi K3.1 Pro leads 4 (BenchLeader, 16 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Either | 59 vs 60 category score | Either |
| Agents & tool use | Kimi K3.1 Pro | 65 vs 58 category score | Medium |
| Maths | Kimi K3.1 Pro | 73 vs 63 category score | High |
| Reasoning | GLM-5 | 68 vs 59 category score; Humanity's Last Exam (AA) 29.3% vs 52.5% | High |
| Knowledge | Either | 57 vs 58 category score | Either |
| Long documents | Kimi K3.1 Pro | 68 vs 62 category score; AA-LCR v1.1 86.3% vs 75.7% | Medium |
| Lowest cost | GLM-5 | $1.55 vs $4.00 per 1M blended; agentic coding session $0.073 vs $0.160 | – |
| Fastest first answer | GLM-5 | 29.2 s (GLM-5 (thinking)) vs 157.4 s (Kimi K3.1 Pro) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GLM-5 (thinking) and Kimi K3.1 Pro are neck and neck on quality (60.2 vs 61.8, within the ±4.7 margin).
- GLM-5 (thinking) is stronger in reasoning.
- Kimi K3.1 Pro is stronger in agents & tools, coding, knowledge, long context, maths.
- GLM-5 (thinking) is 2.6× cheaper ($1.55 vs $4.00 per 1M blended).
- GLM-5 (thinking) streams 2.3× faster (69 vs 30 tokens per second).
At a glance
| Spec | GLM-5 | Kimi K3.1 Pro |
|---|---|---|
| Developer | Zhipu AI | Moonshot AI |
| Released | 11 Feb 2026 | 23 Oct 2026 |
| Context window (tokens) | 204,800 | 1,048,576 |
| Compared at | GLM-5 (thinking) | Kimi K3.1 Pro |
| Open weights | Yes | Yes |
| Price in / out / cached input per 1M | $1.00 / $3.20 / $0.200 | $2.00 / $10.00 / $0.200 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | GLM-5 (thinking) | Kimi K3.1 Pro | Leader |
|---|---|---|---|
| Reasoning | 68.2 | 59.2 | GLM-5 |
| Coding | 59.1 | 59.6 | level |
| Agents & tools | 58.0 | 64.6 | Kimi K3.1 Pro |
| Maths | 63.4 | 73.4 | Kimi K3.1 Pro |
| Knowledge | 57.4 | 58.2 | level |
| Instruction following | 70.9 | – | not comparable |
| Multimodal | – | 66.3 | not comparable |
| Long context | 62.4 | 67.9 | Kimi K3.1 Pro |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Instruction following, Multimodal, Human preference.
Where sources disagree
Across the 5 benchmarks both have, GLM-5 leads 1, Kimi K3.1 Pro leads 4. Every overall index that rates both puts Kimi K3.1 Pro ahead.
| Overall index | GLM-5 (thinking) | Kimi K3.1 Pro | Leader |
|---|---|---|---|
| BenchLeader Index | 60.2 | 61.8 | neck and neck |
| Arena Text | 1457† | 1485 | Kimi K3.1 Pro |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | GLM-5 (thinking) | Kimi K3.1 Pro |
|---|---|---|
| Blended price per 1M tokens | $1.55 | $4.00 |
| Input / output per 1M | $1.00 / $3.20 | $2.00 / $10.00 |
| Cached input per 1M | $0.200 | $0.200 |
| Price per task: chat reply | $0.0014 | $0.0038 |
| Price per task: summarize a 30-page report | $0.014 | $0.030 |
| Price per task: code edit | $0.011 | $0.027 |
| Price per task: agentic coding session | $0.073 | $0.160 |
| Price per task: structured extraction | $0.0026 | $0.0060 |
| Agentic coding session, input mostly cached | $0.037 | $0.079 |
| Cost per benchmark task (AA Index) | – | $3.68 |
| Output tokens per task (AA Index) | – | 72k |
| Of which reasoning | – | 44k |
| Time per task (AA Index) | – | 40 min |
| Output speed | 69 tok/s | 30 tok/s |
| Time to first answer | 29.2 s | 157.4 s |
| Time to a complete answer, last measured 13 Nov 2026 | 81.4 s | 417.5 s |
Each model is compared at its best-ranked reasoning-effort setting. GLM-5 (thinking) scores 60.2 and takes 29 s to its first answer token; GLM-5 scores 54.9 in 2 s. Kimi K3.1 Pro scores 61.8 and takes 157 s to its first answer token; Kimi K3.1 Pro (max) scores 59.9 in 157 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
135 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | GLM-5 (thinking) | Kimi K3.1 Pro |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.2 | 61.8 |
| Agents & tools score | BenchLeader | 58.0 | 64.6 |
| Coding score | BenchLeader | 59.1 | 59.6 |
| Instruction following score | BenchLeader | 70.9 | – |
| Knowledge score | BenchLeader | 57.4 | 58.2 |
| Long context score | BenchLeader | 62.4 | 67.9 |
| Maths score | BenchLeader | 63.4 | 73.4 |
| Reasoning score | BenchLeader | 68.2 | 59.2 |
| Multimodal score | BenchLeader | – | 66.3 |
| Blended price $/M | BenchLeader | $1.55 | $4.00 |
| Output speed | BenchLeader | 69 tok/s | 30 tok/s |
| Time to first answer | BenchLeader | 29.2 s | 157.4 s |
| Context window | BenchLeader | 204,800 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 87.8%† | – |
| SimpleBench | SimpleBench | 53.2%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1478† | 1504† |
| LiveBench 2.0 Reasoning | LiveBench | – | 91.1%† |
| GPQA Diamond (AA) | Artificial Analysis | 82.0% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 29.3% | 52.5% |
| HLE-Diamond (AA) | Artificial Analysis | – | 37.3% |
| GPQA Diamond (Vals) | Vals AI | 83.3% | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 75.0% | – |
| ARC-AGI-1 | ARC Prize | 44.7%† | – |
| ARC-AGI-2 | ARC Prize | 4.9%† | – |
| CritPt | Artificial Analysis | 2.0% | 24.6% |
| CritPt v2 hard subset (AA) | Artificial Analysis | – | 49.5% |
| MysteryMechanism | Vals AI | – | 26.1%† |
| Arena Expert | Arena (LMArena) | 1482† | 1515† |
| CL-bench | Epoch AI Benchmarking Hub | 18.7%† | – |
| LMCA | Epoch AI Benchmarking Hub | – | 57.6%† |
| DTBench | Epoch AI Benchmarking Hub | – | 93.3%† |
| ForecastBench | ForecastBench | 63.4%† | – |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 72.1%† | – |
| SciCode | SciCode | – | 60.8% |
| WeirdML | WeirdML | 48.2%† | – |
| Arena Coding | Arena (LMArena) | 1498† | 1530† |
| Arena WebDev | Arena (LMArena) | 1435† | 1687† |
| LiveBench 2.0 Coding | LiveBench | – | 67.8%† |
Show all 135 benchmarks
| Metric | Publisher | GLM-5 (thinking) | Kimi K3.1 Pro |
|---|---|---|---|
| Coding | |||
| LiveCodeBench | Vals AI | 81.9% | – |
| IOI | Vals AI | – | 81.9%† |
| SWE-bench (Vals) | Vals AI | 71.4% | – |
| SWE-bench Verified (bash only) | SWE-bench | 72.8%† | – |
| SWE-bench Multilingual (bash only) | SWE-bench | 69.7%† | – |
| SWE-bench Verified (any scaffold) | SWE-bench | 72.8%† | – |
| Vibe Code Bench v1.1 | Vals AI | 23.4% | 84.9%† |
| SWE Atlas v1.0: Codebase QnA | Scale AI SEAL | 20.5%† | – |
| SWE Atlas v1.0: Refactoring | Scale AI SEAL | 24.2%† | – |
| SWE Atlas v1.0: Test Writing | Scale AI SEAL | 28.7%† | – |
| SWE Atlas v1.1 | Scale AI SEAL | – | 61.1% |
| SWE Atlas v1.1: Codebase QnA | Scale AI SEAL | – | 70.2% |
| SWE Atlas v1.1: Test Writing | Scale AI SEAL | – | 58.9% |
| SWE Atlas v1.1: Refactoring | Scale AI SEAL | – | 54.3% |
| ALE-Bench | Epoch AI Benchmarking Hub | 765.6† | 1932.8† |
| WeirdML v3 | WeirdML | – | 9.9%† |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | – | 67.2% |
| GameBench 2 Overall | DuelLab GameBench 2 | – | 78.3 |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | – | 84.0† |
| Design Arena Website | Design Arena | 1239† | 1321 |
| FrontierCode 1.1 (AA) | Artificial Analysis | – | 47.1% |
| FrontierCode 2.0 (AA) | Artificial Analysis | – | 25.4% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 52.4%† | 59.7%† |
| APEX-Agents | Mercor | – | 65.5% |
| Arena Agent | Arena (LMArena) | – | 4.9† |
| LiveBench Long Agentic | LiveBench | – | 38.7%† |
| LiveBench Browser | LiveBench | – | 75.9%† |
| Terminal-Bench Hard | Artificial Analysis | 43.2% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 98.3% | – |
| HiL-Bench | Scale AI SEAL | – | 61.3% |
| τ²-bench | τ²-bench | 81.0% | – |
| τ³-bench Banking | τ²-bench | 9.8% | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | – | 67.3% |
| GDPval-AA v2.1 | Artificial Analysis | – | 1861 |
| Analyst Agent (AA) | Artificial Analysis | – | 38.8% |
| APEX-Agents (AA) | Artificial Analysis | 14.4% | – |
| Finance Agent v2.1 | Vals AI | – | 61.5%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 11.7%† |
| Legal Research Bench | Vals AI | – | 37.5%† |
| SREBench | Vals AI | – | 27.1%† |
| Tax Agent Bench | Vals AI | – | 24.9%† |
| Terminal-Bench 2.0 (Vals) | Vals AI | 49.4% | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 63.6%† |
| Terminal-Bench Science | Vals AI | – | 31.4%† |
| Arena Agent Task Completion | Arena (LMArena) | – | 2.7† |
| Vending-Bench 2 | Andon Labs | 4432.1† | – |
| AutomationBench | Artificial Analysis | – | 73.3% |
| GDP.pdf (AA) | Artificial Analysis | – | 31.4% |
| AA-Briefcase v1.1 | Artificial Analysis | – | 1674 |
| Design Arena Agentic | Design Arena | 1115† | 1301 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | – | 34.3% |
| CWE-Bench (AA) | Artificial Analysis | – | 60.8% |
| DeepSec-Bench (AA) | Artificial Analysis | – | 36.7% |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 94.7% |
| MKA (AA) | Artificial Analysis | – | 43.9% |
| Maths | |||
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 80.0%† | – |
| ProofBench | Vals AI | – | 91.0% |
| AIME (Vals) | Vals AI | 91.7% | – |
| AIME 2026 | MathArena | 96.7%† | – |
| HMMT February 2026 | MathArena | 86.4%† | – |
| USAMO 2026 | MathArena | 35.1%† | – |
| MathArena Apex | MathArena | 10.9%† | – |
| ArXivMath | MathArena | – | 58.3%† |
| BrokenArXiv | MathArena | – | 61.9%† |
| Arena Maths | Arena (LMArena) | 1443† | – |
| Knowledge | |||
| LiveBench 2.0 Data Analysis | LiveBench | – | 79.1%† |
| LiveBench Research | LiveBench | – | 82.1%† |
| AA-Omniscience v1.1 | Artificial Analysis | – | 29.9 |
| MMLU-Pro | Vals AI | 86.0% | – |
| LegalBench | Vals AI | 84.1% | – |
| CorpFin | Vals AI | 62.9% | – |
| TaxEval | Vals AI | 70.0% | – |
| MedQA | Vals AI | 94.3% | – |
| PRBench Finance | Scale AI SEAL | – | 47.5% |
| PRBench Legal | Scale AI SEAL | – | 47.9% |
| BioMysteryBench | Vals AI | – | 75.6%† |
| CaseLaw v2 | Vals AI | 52.5% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 70.0%† |
| Access to Justice Benchmark | Vals AI | – | 69.5%† |
| DrugDiscoveryBench | Scale AI SEAL | – | 63.4%† |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | – | 48.9% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | – | 62.7% |
| Global-MMLU-Lite (AA) | Artificial Analysis | 81.9%† | – |
| Instruction following | |||
| IFBench | Artificial Analysis | 72.3% | – |
| Arena Instruction Following | Arena (LMArena) | 1447† | 1487† |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1457† | 1485† |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1601† | 2173 |
| Arena Creative Writing | Arena (LMArena) | 1447† | 1453† |
| Arena Multi-turn | Arena (LMArena) | 1472† | 1475† |
| Arena Longer Queries | Arena (LMArena) | 1469† | 1491† |
| Multimodal | |||
| LiveBench Design | LiveBench | – | 73.7%† |
| MMMU-Pro | Artificial Analysis | – | 83.9% |
| Humanity's Sixth Sense | Scale AI SEAL | – | 33.2%† |
| VIALS-AA | Handshake AI | – | 45.3% |
| AA-3D | Artificial Analysis | – | 1112 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 75.7% | 86.3% |
| MLCR | Artificial Analysis | – | 38.3% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 146.5† | 161.4 |
| LiveBench 2.0 | LiveBench | – | 68.7%† |
| AA Intelligence Index v5.0.1 | Artificial Analysis | – | 46.5 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 27.9 | 55.1 |
| AA Progress Index v1 | Artificial Analysis | 280.9 | 340.7 |
| Vals Index | Vals AI | – | 62.5† |
| Vals Finance Index | Vals AI | – | 56.0%† |
| Vals Science Index | Vals AI | – | 44.4%† |
| Vals Legal Index | Vals AI | – | 67.0%† |
| AA Cyber Index | Artificial Analysis | – | 64.1 |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | – | 18.6% |
| DistressBench | Scale AI SEAL | 71.2%† | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 50 | 33.3 |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Kimi K3.1 Pro: BenchLeader Index 61.2 → 61.8.
- Kimi K3.1 Pro: 1 score revision.
- Kimi K3.1 Pro: 1 score revision.
- Kimi K3.1 Pro: 1 new result (HiL-Bench).
- Kimi K3.1 Pro: 3 new results (PRBench Finance, PRBench Legal, FORTRESS); 2 score revisions.
- Kimi K3.1 Pro: 2 new results (APEX-Agents, Epoch Capabilities Index).
- Kimi K3.1 Pro: BenchLeader Index 61.8 → 61.2.
- Kimi K3.1 Pro: 5 score revisions.
- Kimi K3.1 Pro: BenchLeader Index 61.3 → 61.8.
- GLM-5 (thinking): cached input price moved from $0.15/M to $0.20/M (34% up).
Related comparisons
- GLM-5 vs GPT-6 Astra
- Kimi K3.1 Pro vs Claude Fable 5.2
- Kimi K3.1 Pro vs Gemini 4 Carbon
- Kimi K3.1 Pro vs Claude Opus 5.5
- Kimi K3.1 Pro vs GPT-6 Astra
- GLM-5 vs Kimi K3
- GLM-5 vs GPT-6.1 Sol
- GLM-5 vs Gemini 4 Argon
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
GLM-5 vs Kimi K3.1 Pro: questions
- Is GLM-5 better than Kimi K3.1 Pro?
- GLM-5 (thinking) and Kimi K3.1 Pro are neck and neck on quality (60.2 vs 61.8, within the ±4.7 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
- GLM-5 vs Kimi K3.1 Pro: which is cheaper, and will that change?
- GLM-5 is cheaper today: $1.55 against $4.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is GLM-5 or Kimi K3.1 Pro better for coding?
- They are level in coding (59 vs 60 on the category index).
- Which is faster, GLM-5 or Kimi K3.1 Pro?
- GLM-5 streams ~2× faster (69 vs 30 output tokens per second, Artificial Analysis 7-day median).
- Is GLM-5 or Kimi K3.1 Pro better for agents and tool use?
- Kimi K3.1 Pro scores higher in agentic tasks (65 vs 58 on the category index, where 50 is average).
More questions (4)
- Which is better for long documents?
- Kimi K3.1 Pro scores higher on long-context benchmarks (68 vs 62). Kimi K3.1 Pro accepts more context: 1,048,576 against 204,800 tokens for GLM-5.
- Which has lower latency?
- GLM-5 answers first: 29.2 s against 157.4 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on GLM-5 vs Kimi K3.1 Pro?
- Across the 5 benchmarks both have, GLM-5 leads 1, Kimi K3.1 Pro leads 4. Every overall index that rates both puts Kimi K3.1 Pro ahead.
- When was Kimi K3.1 Pro released?
- Kimi K3.1 Pro was released on 23 Oct 2026.