Zhipu AIGLM 4.6V
GLM 4.6V is a Zhipu AI open-weights reasoning model, released 8 Dec 2025. Its best configuration (thinking reasoning effort) ranks #544 of 785 on the BenchLeader Index at 44.2 ±4.9, in the lower half. The ± is the point: 313 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in knowledge (49) and lowest in multimodal (31). At $0.450 per million tokens blended it is cheaper than most ranked models. Output speed of 89 tokens per second puts it faster than most, with a first answer in 26.0 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 8 Dec 2025.stale: no new result in six months
- Blended price
- $0.450/M
- $0.300 in · $0.900 out
- Output speed
- 89 tok/s
- measured by Artificial Analysis
- First answer
- 26 s
- first token 2.82 s
- Context
- 128k
- Full answer
- 32 s
- median, reasoning included
- Cost per run
- –
- Released
- 8 Dec 2025
- Overall index44
- Reasoning41
- Agents & tools46
- Knowledge49
- Instruction following34
- Multimodal31
- Long context49
- Composite42
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 39.9 | #669 | 79 tok/s | 3.68 s | $0.0004 | Agents & tools 36 · Composite 39 · Instruction following 32 · Knowledge 45 · Long context 32 · Multimodal 24 · Reasoning 41 |
| thinkingbest | 44.2 | #544 | 89 tok/s | 26 s | $0.0004 | Agents & tools 46 · Composite 42 · Instruction following 34 · Knowledge 49 · Long context 49 · Multimodal 31 · Reasoning 41 |
| not stated | – | – | 28 tok/s | 2.82 s | $0.0002 | Coding 54 · Human preference 55 · Multimodal 47 · Reasoning 54 |
How its index has moved
9 Sept 2026 to 21 Oct 2026- thinking
- no reasoning
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena Hard Promptseffort not stated | – | – | 1384#185 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 56.6%#399 | 71.9%#282 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 3.7%#577 | 9.6%#332 | – | Artificial Analysis | |
| CritPt | 0.0%#315 | 0.0%#315 | – | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1415#183 | Arena (LMArena) |
Agents & tools
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Terminal-Bench Hard | 3.0%#314 | 14.4%#196 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 30.7%#255 | 31.6%#251 | Artificial Analysis |
Knowledge
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-Omniscience | -37.6#313 | -26.9#269 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 17.4%#367 | 16.2%#397 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 33.4%#200 | 48.5%#125 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| IFBench | 27.9%#376 | 30.1%#364 | – | Artificial Analysis | |
| Arena Instruction Followingnot in indexeffort not stated | – | – | 1366#175 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1378#176 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1351#167 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1367#182 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1381#176 | Arena (LMArena) |
Multimodal
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena Visioneffort not stated | – | – | 1164#110 | Arena (LMArena) | |
| MMMU-Pro | 42.2%#262 | 48.5%#244 | – | Artificial Analysis |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 17.0%#446 | 48.7%#329 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 8.3#440 | 11.2#357 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| NovitaAI | 36 tok/s | 2.82 s | $0.300 | $0.900 | 131k | bf16 |
| Z.ai | 32 tok/s | 2.96 s | $0.300 | $0.900 | 131k | fp8 |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.030 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0004 | $0.0003 | 29.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0041 | $0.0017 | 32.7 s |
| Code edit | 6,000 / 1,500 | $0.0032 | $0.0019 | 42.8 s |
| Agentic coding session | 60,000 / 4,000 | $0.022 | $0.0094 | 1.2 min |
| Structured extraction | 2,000 / 200 | $0.0008 | $0.0004 | 28.3 s |
More from Zhipu AI
See also
Questions people ask
- What are GLM 4.6V's benchmark scores?
- As of 25 Oct 2026, GLM 4.6V (thinking effort) scores 44.2 ±4.9 on the BenchLeader Index, strongest in knowledge (49) and weakest in multimodal (31). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does GLM 4.6V rank?
- GLM 4.6V (thinking effort) is #544 of 785 ranked configurations on 25 Oct 2026. 313 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does GLM 4.6V cost per 1M tokens?
- As of 25 Oct 2026, GLM 4.6V costs $0.300 per 1M input tokens and $0.900 per 1M output tokens, $0.450 blended (three input tokens per output token).
- When was GLM 4.6V released?
- Zhipu AI released GLM 4.6V on 8 Dec 2025. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GLM 4.6V: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/glm-4-6v.