Zhipu AIGLM 4.7 Flash
GLM 4.7 Flash is a Zhipu AI open-weights model, released 19 Jan 2026. Its best configuration ranks #349 of 785 on the BenchLeader Index at 51.3 ±2.7, in the upper half. The ± is the point: 294 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 coding (55) and lowest in reasoning (48). At $0.153 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 43 tokens per second puts it in the slowest quarter, with a first token in 0.4 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 22 Oct 2026.
- Blended price
- $0.153/M
- $0.070 in · $0.400 out
- Output speed
- 43 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 0.39 s
- Context
- 200k
- Full answer
- –
- Cost per run
- –
- Released
- 19 Jan 2026
- Overall index51
- Reasoning48
- Coding55
- Maths52
- Human preference53
Versions
Zhipu AI has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| GLM 5.3 Flash | 20 Aug 2026 | 62.6 | #80 |
| GLM 4.7 Flashthis page | 19 Jan 2026 | 51.3 | #349 |
| GLM-4.5-Flash | 28 Jul 2025 | – | – |
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 | 41.0 | #644 | 99 tok/s | 1.73 s | $0.0001 | Agents & tools 37 · Composite 41 · Instruction following 48 · Knowledge 30 · Long context 34 · Maths 34 · Reasoning 38 |
| thinking | 47.7 | #451 | 86 tok/s | 24 s | $0.0001 | Agents & tools 53 · Composite 46 · Instruction following 61 · Knowledge 33 · Long context 45 · Reasoning 41 |
| not statedbest | 51.3 | #349 | 43 tok/s | 0.39 s | $0.0001 | Coding 55 · Human preference 53 · Maths 52 · Reasoning 48 |
How its index has moved
8 Sept 2026 to 21 Oct 2026- not stated
- 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 |
|---|---|---|---|---|---|
| GPQA Diamond | 45.1%#237 | – | 60.5%#192 | Epoch AI Benchmarking Hub | |
| Arena Hard Promptseffort not stated | – | – | 1384#184 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 45.1%#452 | 58.1%#387 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 5.0%#459 | 7.6%#360 | – | Artificial Analysis | |
| CritPt | 0.0%#315 | 0.3%#271 | – | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1423#179 | Arena (LMArena) |
Agents & tools
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Terminal-Bench Hard | 3.8%#298 | 22.0%#156 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 91.8%#56 | 98.8%#3 | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| OTIS Mock AIME | 25.0%#216 | 58.3%#166 | Epoch AI Benchmarking Hub | |
| Arena Mathseffort not stated | – | 1362#188 | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-Omniscience | -68.9#508 | -62.6#482 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 13.1%#463 | 16.1%#399 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 5.7%#515 | 6.1%#510 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| IFBench | 46.3%#194 | 60.8%#124 | – | Artificial Analysis | |
| Arena Instruction Followingnot in indexeffort not stated | – | – | 1348#189 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1365#186 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1305#210 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1359#186 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1369#189 | Arena (LMArena) |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 20.3%#432 | 41.7%#352 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 10.6#375 | 14.9#290 | 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 | 51 tok/s | 1.47 s | $0.070 | $0.400 | 200k | bf16 |
| Cloudflare | 46 tok/s | 0.37 s | $0.060 | $0.400 | 131k | – |
Price history
The price above, per 1M tokens, every day it changed.
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 $0.010 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | $0.0001 | 7.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0011 | $0.0005 | 14.3 s |
| Code edit | 6,000 / 1,500 | $0.0010 | $0.0008 | 35.3 s |
| Agentic coding session | 60,000 / 4,000 | $0.0058 | $0.0031 | 1.6 min |
| Structured extraction | 2,000 / 200 | $0.0002 | $0.0001 | 5.0 s |
See also
Questions people ask
- What are GLM 4.7 Flash's benchmark scores?
- As of 24 Oct 2026, GLM 4.7 Flash scores 51.3 ±2.7 on the BenchLeader Index, strongest in coding (55) and weakest in reasoning (48). That rests on 6 independent benchmark results, and every score on this page links to its publisher.
- How does GLM 4.7 Flash rank?
- GLM 4.7 Flash is #349 of 785 ranked configurations on 24 Oct 2026. 294 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 4.7 Flash cost per 1M tokens?
- As of 24 Oct 2026, GLM 4.7 Flash costs $0.070 per 1M input tokens and $0.400 per 1M output tokens, $0.153 blended (three input tokens per output token). The list price last changed on 9 Sept 2026.
- When was GLM 4.7 Flash released?
- Zhipu AI released GLM 4.7 Flash on 19 Jan 2026. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GLM 4.7 Flash: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/glm-4-7-flash.