Gemini 4 Carbon vs GLM 5.3
Full profiles: Gemini 4 Carbon (Google) and GLM 5.3 (Zhipu AI): every benchmark, price and rank.
Updated · Data from 8 sources
As of 29 Oct 2026, Gemini 4 Carbon and GLM 5.3 are neck and neck on quality (65.3 vs 58.9 on the BenchLeader Index, inside the ±6.8 margin), and GLM 5.3 is 3.7× cheaper at $2.15 vs $8.00 per 1M tokens blended. Gemini 4 Carbon leads in every category we measure both in. Gemini 4 Carbon streams faster.
Key facts (as of 29 Oct 2026)
- Gemini 4 Carbon scores 65.3 on the BenchLeader Index and GLM 5.3 scores 58.9 (BenchLeader, 29 Oct 2026).
- Gemini 4 Carbon costs $8.00 per 1M tokens blended (list price, 29 Oct 2026).
- GLM 5.3 costs $2.15 per 1M tokens blended (list price, 29 Oct 2026).
- Gemini 4 Carbon streams faster than GLM 5.3 (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 26 benchmarks both models have, Gemini 4 Carbon leads 25 (BenchLeader, 29 Oct 2026).
Gemini 4 Carbon benchmarks 
GLM 5.3 benchmarks 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 | Gemini 4 Carbon | 72 vs 60 category score; FrontierCode 2.0 (AA) 40.6% vs 21.6% | High |
| Reasoning | Gemini 4 Carbon | 79 vs 55 category score; HLE-Diamond (AA) 69.1% vs 17.9% | High |
| Knowledge | Gemini 4 Carbon | 76 vs 54 category score; AA-Omniscience v1.1 53.5 vs 6.0 | High |
| Long documents | Either | 65 vs 65 category score; AA-LCR v1.1 79.3% vs 79.7% | Either |
| Chat quality (human preference) | – | Only GLM 5.3 has an Arena rating (67 category score) | Not comparable |
| Lowest cost | GLM 5.3 | $2.15 vs $8.00 per 1M blended; agentic coding session $0.102 vs $0.320 | – |
| Fastest first answer | GLM 5.3 | 30.0 s (GLM 5.3 (max)) vs 103.1 s (Gemini 4 Carbon (high)) 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
- Gemini 4 Carbon (high) and GLM 5.3 (max) are neck and neck on quality (65.3 vs 58.9, within the ±6.8 margin).
- Gemini 4 Carbon (high) is stronger in coding, knowledge, reasoning.
- GLM 5.3 (max) is stronger in long context, agents & tools, human preference, maths.
- GLM 5.3 (max) is 3.7× cheaper ($2.15 vs $8.00 per 1M blended).
- Gemini 4 Carbon (high) streams 1.4× faster (98 vs 72 tokens per second).
At a glance
| Spec | Gemini 4 Carbon | GLM 5.3 |
|---|---|---|
| Developer | Zhipu AI | |
| Released | 28 Oct 2026 | 18 Aug 2026 |
| Context window (tokens) | 1,000,000 | 1,000,000 |
| Compared at | Gemini 4 Carbon (high) | GLM 5.3 (max) |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | $4.00 / $20.00 / – | $1.40 / $4.40 / $0.299 |
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 | Gemini 4 Carbon (high) | GLM 5.3 (max) | Leader |
|---|---|---|---|
| Reasoning | 79.3 | 55.1 | Gemini 4 Carbon |
| Coding | 71.9 | 59.5 | Gemini 4 Carbon |
| Agents & tools | – | 58.0 | not comparable |
| Maths | – | 60.3 | not comparable |
| Knowledge | 76.1 | 53.6 | Gemini 4 Carbon |
| Human preference | – | 67.2 | not comparable |
| Long context | 64.5 | 64.7 | level |
Where sources disagree
Across the 26 benchmarks both have, Gemini 4 Carbon leads 25, GLM 5.3 leads 1. Every overall index that rates both puts Gemini 4 Carbon ahead.
| Overall index | Gemini 4 Carbon (high) | GLM 5.3 (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.3 | 58.9 | neck and neck |
| AA Intelligence Index v5.0 | 59.4 | 31.6 | Gemini 4 Carbon |
| Vals Index | 70.4 | 59.7 | Gemini 4 Carbon |
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; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Gemini 4 Carbon (high) | GLM 5.3 (max) |
|---|---|---|
| Blended price per 1M tokens | $8.00 | $2.15 |
| Input / output per 1M | $4.00 / $20.00 | $1.40 / $4.40 |
| Cached input per 1M | – | $0.299 |
| Price per task: chat reply | $0.0076 | $0.0019 |
| Price per task: summarize a 30-page report | $0.060 | $0.019 |
| Price per task: code edit | $0.054 | $0.015 |
| Price per task: agentic coding session | $0.320 | $0.102 |
| Price per task: structured extraction | $0.012 | $0.0037 |
| Cost to run a full benchmark suite | $7.76 | $5.01 |
| Output speed | 98 tok/s | 72 tok/s |
| Time to first answer | 103.1 s | 30.0 s |
| Time to a complete answer | 108.2 s | 36.9 s |
Per token is not per task: on Artificial Analysis' full suite Gemini 4 Carbon (high) cost $7.76 and GLM 5.3 (max) $5.01, because GLM 5.3 (max) used about 2.4× as many tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. Gemini 4 Carbon (high) scores 65.3 and takes 103 s to its first answer token. GLM 5.3 (max) scores 58.9 and takes 30 s to its first answer token; GLM 5.3 (low) scores 46.8 in 29 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
115 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 | Gemini 4 Carbon (high) | GLM 5.3 (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.3 | 58.9 |
| Coding score | BenchLeader | 71.9 | 59.5 |
| Knowledge score | BenchLeader | 76.1 | 53.6 |
| Long context score | BenchLeader | 64.5 | 64.7 |
| Reasoning score | BenchLeader | 79.3 | 55.1 |
| Agents & tools score | BenchLeader | – | 58.0 |
| Human preference score | BenchLeader | – | 67.2 |
| Maths score | BenchLeader | – | 60.3 |
| Blended price $/M | BenchLeader | $8.00 | $2.15 |
| Output speed | BenchLeader | 98 tok/s | 72 tok/s |
| Time to first answer | BenchLeader | 103.1 s | 30.0 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,000,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | – | 90.9% |
| Arena Hard Prompts | Arena (LMArena) | – | 1505 |
| LiveBench 1.0 Reasoning | LiveBench | – | 85.8%† |
| GPQA Diamond (AA) | Artificial Analysis | – | 91.7% |
| Humanity's Last Exam (AA) | Artificial Analysis | 60.4% | 42.3% |
| HLE-Diamond (AA) | Artificial Analysis | 69.1% | 17.9% |
| GPQA Diamond (Vals) | Vals AI | – | 88.1% |
| CritPt | Artificial Analysis | 36.0% | 19.1% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 63.0% | 33.5% |
| MysteryMechanism | Vals AI | – | 23.0% |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 37.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | – | 33.0% |
| LMCA | Epoch AI Benchmarking Hub | – | 55.5%† |
| DTBench | Epoch AI Benchmarking Hub | – | 87.7%† |
| BTF-3 | Epoch AI Benchmarking Hub | – | 14.9%† |
| Coding | |||
| SciCode | SciCode | 64.2% | 59.0% |
| WeirdML | WeirdML | – | 75.4% |
| FrontierCode | Cognition | – | 40.1% |
| Arena Coding | Arena (LMArena) | – | 1521 |
| Arena WebDev | Arena (LMArena) | – | 1623 |
| LiveBench 1.0 Coding | LiveBench | – | 79.0%† |
| LiveCodeBench | Vals AI | – | 80.5% |
| IOI | Vals AI | – | 68.4% |
| SWE-bench (Vals) | Vals AI | – | 95.4% |
| Code Migration | Vals AI | – | 44.2% |
Show all 115 benchmarks
| Metric | Publisher | Gemini 4 Carbon (high) | GLM 5.3 (max) |
|---|---|---|---|
| Coding | |||
| Code Migration v1.1 | Vals AI | – | 55.1% |
| ProgramBench | Vals AI | – | 1.5% |
| Vibe Code Bench 1-100 | Vals AI | – | 20.0% |
| Vibe Code Bench v1.1 | Vals AI | 89.1% | 78.1% |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | – | 69.0% |
| CursorBench | Cursor | – | 42.6% |
| ALE-Bench | Epoch AI Benchmarking Hub | – | 1317.4† |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 30.2% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 65.2% | 41.9% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 83.9% |
| Startrise Frontend | Startrise | – | 73.2%† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 48.5% | 40.1% |
| FrontierCode 2.0 (AA) | Artificial Analysis | 40.6% | 21.6% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | – | 41.8% |
| APEX-Agents | Mercor | – | 56.6%† |
| Arena Agent | Arena (LMArena) | – | 1.2 |
| LiveBench Agentic Coding | LiveBench | – | 60.9%† |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 71.5% |
| MCP Atlas | Scale AI SEAL | – | 84.2%† |
| HiL-Bench | Scale AI SEAL | – | 58.3%† |
| GDPval-AA v2.1, normalised score | Artificial Analysis | – | 57.5% |
| GDPval-AA v2.1 | Artificial Analysis | 1733 | 1658 |
| τ³-Banking (AA) | Artificial Analysis | – | 50.3% |
| ITBench SRE (AA) | Artificial Analysis | – | 46.1% |
| Analyst Agent (AA) | Artificial Analysis | – | 38.8% |
| CyberBench | Vals AI | – | 72.1% |
| Finance Agent v2 | Vals AI | – | 55.8% |
| Finance Agent v2.1 | Vals AI | 67.2% | 65.0% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 17.5% | 14.6% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 8.3% |
| Legal Research Bench | Vals AI | 55.8% | 49.0% |
| SkillsBench | Vals AI | – | 47.5% |
| SREBench | Vals AI | – | 9.2% |
| Tax Agent Bench | Vals AI | 48.9% | 39.7% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 63.1% | 38.9% |
| Terminal-Bench Science | Vals AI | – | 5.7% |
| Vending-Bench 2 | Andon Labs | 16484.2† | 8163.6† |
| AutomationBench | Artificial Analysis | 77.8% | 62.2% |
| GDP.pdf (AA) | Artificial Analysis | 28.4% | 11.2% |
| Harvey LAB-AA v1.1 | Harvey | – | 0.3% |
| EnterpriseOps-Gym | Artificial Analysis | – | 36.4% |
| AA-Briefcase v1.1 | Artificial Analysis | 1609 | 1504 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 64.8% | 9.5% |
| CWE-Bench (AA) | Artificial Analysis | – | 55.8% |
| DeepSec-Bench (AA) | Artificial Analysis | – | 24.3% |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 29.0% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 68.8% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 29.3% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | – | 91.1% |
| ProofBench | Vals AI | – | 49.0% |
| LiveBench Mathematics | LiveBench | – | 87.9%† |
| Arena Maths | Arena (LMArena) | – | 1500 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | – | 41.0% |
| LiveBench 1.0 Data Analysis | LiveBench | – | 70.2%† |
| AA-Omniscience v1.1 | Artificial Analysis | 53.5 | 6.0 |
| MMLU-Pro | Vals AI | – | 86.8% |
| LegalBench | Vals AI | – | 84.8% |
| TaxEval | Vals AI | – | 72.4% |
| Excel Modeling Benchmark | Vals AI | – | 56.3% |
| Excel Modeling Benchmark v1.1 | Vals AI | 76.5% | 69.1% |
| MedCode | Vals AI | – | 42.9% |
| MedScribe | Vals AI | – | 88.8% |
| Public Benefits Bench v1.1 | Vals AI | – | 68.5% |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 58.3% | 32.3% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 88.6% | 61.1% |
| HealthBench Professional (AA) | OpenAI | – | 52.8% |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 79.9%† |
| LiveBench Instruction Following | LiveBench | – | 69.3%† |
| Arena Instruction Following | Arena (LMArena) | – | 1478 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1481 |
| Arena Creative Writing | Arena (LMArena) | – | 1456 |
| Arena Multi-turn | Arena (LMArena) | – | 1478 |
| Arena Longer Queries | Arena (LMArena) | – | 1486 |
| Multimodal | |||
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 58.6%† | – |
| Code2Video-AA | Artificial Analysis | – | 1034 |
| AA-3D | Artificial Analysis | – | 783 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.3% | 79.7% |
| MLCR | Artificial Analysis | – | 17.2% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | – | 156.0† |
| LiveBench 1.0 | LiveBench | – | 76.1%† |
| AA Intelligence Index v5.0 | Artificial Analysis | 59.4 | 31.6 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | – | 44.8 |
| AA Progress Index v1 | Artificial Analysis | 351.6 | 322.7 |
| Vals Index | Vals AI | 70.4 | 59.7 |
| Vals Finance Index | Vals AI | 66.3% | 60.5% |
| Vals RSI Index | Vals AI | – | 22.9% |
| Vals Router Index | Vals AI | – | 59.7%† |
| AA Cyber Index | Artificial Analysis | – | 36.4 |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | – | 28.2% |
| Metadata | |||
| AA Openness Index | Artificial Analysis | – | 33.3 |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- GLM 5.3 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
- GLM 5.3 (max): BenchLeader Index 61.3 → 58.9.
- GLM 5.3 (max): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $2.01/M to $5.01/M (150% up); cached input price moved from $0.30/M to $0.26/M (13% down).
- GLM 5.3 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
- GLM 5.3 (max): cached input price moved from $0.30/M to $0.26/M (13% down).
- GLM 5.3 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
- GLM 5.3 (max): cached input price moved from $0.30/M to $0.26/M (13% down).
- GLM 5.3 (max): 6 new results (Code Migration v1.1, Excel Modeling Benchmark v1.1, Finance Agent v2.1…).
- GLM 5.3 (max): 1 score revision.
- GLM 5.3 (max): 4 score revisions; cached input price moved from $0.26/M to $0.30/M (15% up).
Related comparisons
- Gemini 4 Carbon vs Claude Opus 5.5
- Gemini 4 Carbon vs GPT-6 Astra
- Gemini 4 Carbon vs GPT-6.1 Sol
- GLM 5.3 vs Claude Opus 5.5
- GLM 5.3 vs GPT-6 Astra
- GLM 5.3 vs GPT-6.1 Sol
- GLM 5.3 vs Claude Fable 5.1
- Gemini 4 Argon vs GLM 5.3
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Gemini 4 Carbon vs GLM 5.3: questions
- Is Gemini 4 Carbon better than GLM 5.3?
- Gemini 4 Carbon (high) and GLM 5.3 (max) are neck and neck on quality (65.3 vs 58.9, within the ±6.8 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 29 Oct 2026, so check the category scores for your use.
- Gemini 4 Carbon vs GLM 5.3: which is cheaper, and will that change?
- GLM 5.3 is cheaper today: $2.15 against $8.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Gemini 4 Carbon or GLM 5.3 better for coding?
- Gemini 4 Carbon scores higher in coding (72 vs 60 on the category index, where 50 is average). On FrontierCode 2.0 (AA), the top shared coding benchmark, it is 40.6% for Gemini 4 Carbon against 21.6% for GLM 5.3.
- Which is faster, Gemini 4 Carbon or GLM 5.3?
- Gemini 4 Carbon streams faster (98 vs 72 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- They are level on long-context benchmarks (65 vs 65). Both accept 1,000,000 tokens of context.
More questions (3)
- Which has lower latency?
- GLM 5.3 answers first: 30.0 s against 103.1 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 Gemini 4 Carbon vs GLM 5.3?
- Across the 26 benchmarks both have, Gemini 4 Carbon leads 25, GLM 5.3 leads 1. Every overall index that rates both puts Gemini 4 Carbon ahead.
- When was Gemini 4 Carbon released?
- Gemini 4 Carbon was released on 28 Oct 2026.