GLM-4.7 vs Kimi K2 Thinking vs Nemotron 3 Ultra 550B A55B
Nemotron 3 Ultra 550B A55B comes out ahead, 62 to 58 and 56 on our weighted score.
Z.ai (Zhipu)GLM-4.7
56/100- ECI143.5
- Price$0.60 / $2.20
- Context205K
Moonshot AIKimi K2 Thinking
58/100- ECI146.0
- Price$0.60 / $2.50
- Context262K
- Our pick
NVIDIANemotron 3 Ultra 550B A55B
62/100- ECI146.2
- Price$0.50 / $2.50
- Context1M
Nemotron 3 Ultra 550B A55B is our pick
Nemotron 3 Ultra 550B A55B is the better all-round choice, scoring 62/100 against Kimi K2 Thinking (58) and GLM-4.7 (56). It leads on inputs & features and context window. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityNemotron 3 Ultra 550B A55BCapabilities Index (ECI): Nemotron 3 Ultra 550B A55B 146.2 · Kimi K2 Thinking 146.0 · GLM-4.7 143.5
- Lowest priceGLM-4.7 and Nemotron 3 Ultra 550B A55BGLM-4.7 $1.00 · Nemotron 3 Ultra 550B A55B $1.00 · Kimi K2 Thinking $1.07 per 1M tokens (3:1 blend)
- Longest contextNemotron 3 Ultra 550B A55BNemotron 3 Ultra 550B A55B 1,000,000 · Kimi K2 Thinking 262,144 · GLM-4.7 204,800 tokens
- Widest inputsSame inputsGLM-4.7: Text · Kimi K2 Thinking: Text · Nemotron 3 Ultra 550B A55B: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | GLM-4.7 | Kimi K2 Thinking | Nemotron 3 Ultra 550B A55B |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 70 | 73 | 73 |
| Price | 25% | 50 | 48 | 50 |
| Inputs & features | 15% | 35 | 35 | 45 |
| Context window | 10% | 32 | 37 | 60 |
| Overall | 100% | 56/100 | 58/100 | 62/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | GLM-4.7Z.ai (Zhipu) | Kimi K2 ThinkingMoonshot AI | Nemotron 3 Ultra 550B A55BNVIDIA |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 143.5 | 146.0 | 146.2 (best) |
| ECI rank | #84 of 148 | #72 of 148 | #70 of 148 (best) |
| GPQA DiamondGraduate-level science questions | 83.3% | 84.2% | 85.4% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 83.3% | 83.1% | 86.7% (best) |
| SimpleQA VerifiedShort factual questions | 32.2% | — | — |
| Price per million tokens | |||
| Input | $0.60 | $0.60 | $0.50 (best) |
| Output | $2.20 (best) | $2.50 | $2.50 |
| Cached input | $0.11 (best) | — | $0.15 |
| Blended (3:1) | $1.00 (best) | $1.07 | $1.00 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Z.AI API | Median of 10 providers | Official Nvidia API |
| Limits | |||
| Context window | 204,800 tokens | 262,144 tokens | 1,000,000 tokens (best) |
| Max output | 131,072 tokens | 262,144 tokens (best) | 128,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | Yes |
| Availability | |||
| Weights | Open | Open | Open |
| API model ID | glm-4.7 | — | nvidia/nemotron-3-ultra-550b-a55b |
| API providers | 20 | 10 | 21 (best) |
| Released | Dec 22, 2025 | Nov 6, 2025 | Jun 4, 2026 |
| Knowledge cutoff | Apr 2025 | Aug 2024 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- GLM-4.7$10.40
- Kimi K2 Thinking$11.00
- Nemotron 3 Ultra 550B A55B$10.00
Which should you choose?
Which is better: GLM-4.7, Kimi K2 Thinking or Nemotron 3 Ultra 550B A55B?
Nemotron 3 Ultra 550B A55B is the better all-round choice, scoring 62/100 against Kimi K2 Thinking (58) and GLM-4.7 (56). It leads on inputs & features and context window. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, GLM-4.7, Kimi K2 Thinking or Nemotron 3 Ultra 550B A55B?
GLM-4.7 is cheaper at $0.60 input / $2.20 output per million tokens (official Z.AI API price). Nemotron 3 Ultra 550B A55B costs $0.50 input / $2.50 output per million tokens (official Nvidia API price); Kimi K2 Thinking costs $0.60 input / $2.50 output per million tokens (median across 10 API providers). At a typical mix of three input tokens to one output token, that is $1.00 per million tokens for GLM-4.7 versus $1.00 for Nemotron 3 Ultra 550B A55B (1× as much) and $1.07 for Kimi K2 Thinking (1.1× as much).
Which scores higher on benchmarks?
Nemotron 3 Ultra 550B A55B scores higher on the Capabilities Index (ECI): Nemotron 3 Ultra 550B A55B 146.2 (#70 of 148), Kimi K2 Thinking 146.0 (#72 of 148) and GLM-4.7 143.5 (#84 of 148). The confidence ranges of the top two overlap (143.9–148.1 vs 143.4–147.6), so treat the gap as small. On individual benchmarks: GPQA Diamond — Nemotron 3 Ultra 550B A55B 85.4%, Kimi K2 Thinking 84.2%, GLM-4.7 83.3%; OTIS Mock AIME 2024–2025 — Nemotron 3 Ultra 550B A55B 86.7%, GLM-4.7 83.3%, Kimi K2 Thinking 83.1%.
Which is better for coding?
There are no published SWE-bench Verified results for GLM-4.7, Kimi K2 Thinking and Nemotron 3 Ultra 550B A55B yet, so there is no like-for-like coding score. On overall capability, Nemotron 3 Ultra 550B A55B leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.
Which has the bigger context window?
Nemotron 3 Ultra 550B A55B has the largest context window at 1,000,000 tokens, against 262,144 for Kimi K2 Thinking and 204,800 for GLM-4.7. Maximum output per response: GLM-4.7 up to 131,072, Kimi K2 Thinking up to 262,144, Nemotron 3 Ultra 550B A55B up to 128,000 tokens.
Which can read images, PDFs, audio or video?
GLM-4.7 accepts text; Kimi K2 Thinking accepts text; Nemotron 3 Ultra 550B A55B accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights, so you can self-host them.
Which is newer?
Nemotron 3 Ultra 550B A55B is the newest, released Jun 4, 2026. GLM-4.7 came out Dec 22, 2025; Kimi K2 Thinking came out Nov 6, 2025. Knowledge cutoff: GLM-4.7 Apr 2025, Kimi K2 Thinking Aug 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
Related comparisons
- GLM-4.7 vs Qwen3.5 35B-A3BAlibaba (Qwen) · ECI 142.5
- GLM-4.7 vs Claude Haiku 4.5Anthropic · ECI 142.4
- Kimi K2 Thinking vs Qwen3.5 PlusAlibaba (Qwen) · ECI 146.8
- Kimi K2 Thinking vs GPT-5 MiniOpenAI · ECI 145.5
- Nemotron 3 Ultra 550B A55B vs Qwen3.6 27BAlibaba (Qwen) · ECI 146.5
- Nemotron 3 Ultra 550B A55B vs Gemini 3.5 Flash LiteGoogle · ECI 145.1
