Kimi K2 Thinking vs Llama-3.1-70B-Instruct vs WizardLM 2 8x22B
Kimi K2 Thinking comes out ahead, 63 to 43 and 39 on our weighted score, though WizardLM 2 8x22B is 2.2× cheaper per token.
Price in rating: Included · Exclude price
- Our pick
Moonshot AI
Kimi K2 Thinking
63/100- ECI145.5
- Price$0.60 / $2.50
- Context262K
Meta
Llama-3.1-70B-Instruct
43/100- ECI124.7
- Price$0.72 / $0.72
- Context128K
Microsoft
WizardLM 2 8x22B
39/100- ECI—
- Price$0.493 / $0.493
- Context66K
Kimi K2 Thinking is our pick
Kimi K2 Thinking is the better all-round choice, scoring 63/100 against Llama-3.1-70B-Instruct (43) and WizardLM 2 8x22B (39). It leads on capability, inputs & features and context window. WizardLM 2 8x22B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond), because WizardLM 2 8x22B has no Capabilities Index score yet.
- CapabilityKimi K2 ThinkingShared benchmarks: Kimi K2 Thinking 84.2% · Llama-3.1-70B-Instruct 44.2% · WizardLM 2 8x22B 43.4%
- Lowest priceWizardLM 2 8x22BWizardLM 2 8x22B $0.493 · Llama-3.1-70B-Instruct $0.72 · Kimi K2 Thinking $1.07 per 1M tokens (3:1 blend)
- Longest contextKimi K2 ThinkingKimi K2 Thinking 262,144 · Llama-3.1-70B-Instruct 128,000 · WizardLM 2 8x22B 65,536 tokens
- Widest inputsSame inputsKimi K2 Thinking: Text · Llama-3.1-70B-Instruct: Text · WizardLM 2 8x22B: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Kimi K2 Thinking | Llama-3.1-70B-Instruct | WizardLM 2 8x22B |
|---|---|---|---|---|
| CapabilityShared benchmarks | 50% | 84 | 44 | 43 |
| Price | 25% | 46 | 54 | 62 |
| Inputs & features | 15% | 35 | 25 | 0 |
| Context window | 10% | 43 | 32 | 22 |
| Overall | 100% | 63/100 | 43/100 | 39/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 145.5 (best) | 124.7 | — |
| ECI rank | #90 of 158 (best) | #146 of 158 | — |
| GPQA DiamondGraduate-level science questions | 84.2% (best) | 44.2% | 43.4% |
| OTIS Mock AIME 2024–2025Competition mathematics | 83.1% (best) | 3.6% | — |
| Price per million tokens | |||
| Input | $0.60 | $0.72 | $0.493 (best) |
| Output | $2.50 | $0.72 | $0.493 (best) |
| Cached input | — | — | — |
| Blended (3:1) | $1.07 | $0.72 | $0.493 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 14 providers | Median of 7 providers | Median of 3 providers |
| Limits | |||
| Context window | 262,144 tokens (best) | 128,000 tokens | 65,536 tokens |
| Max output | 262,144 tokens (best) | 4,096 tokens | 8,000 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 | No | No |
| Tool calling | Yes | Yes | No |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Open | OpenApache-2.0 |
| API model ID | — | — | — |
| API providers | 15 (best) | 7 | 3 |
| Released | Nov 6, 2025 | Jul 23, 2024 | Apr 15, 2024 |
| Knowledge cutoff | Aug 2024 | Dec 2023 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Kimi K2 Thinking$11.00
Llama-3.1-70B-Instruct$8.64
WizardLM 2 8x22B$5.92
Which should you choose?
Which is better: Kimi K2 Thinking, Llama-3.1-70B-Instruct or WizardLM 2 8x22B?
Kimi K2 Thinking is the better all-round choice, scoring 63/100 against Llama-3.1-70B-Instruct (43) and WizardLM 2 8x22B (39). It leads on capability, inputs & features and context window. WizardLM 2 8x22B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond), because WizardLM 2 8x22B has no Capabilities Index score yet.
Which is cheaper, Kimi K2 Thinking, Llama-3.1-70B-Instruct or WizardLM 2 8x22B?
WizardLM 2 8x22B is cheaper at $0.493 input / $0.493 output per million tokens (median across 3 API providers). Llama-3.1-70B-Instruct costs $0.72 input / $0.72 output per million tokens (median across 7 API providers); Kimi K2 Thinking costs $0.60 input / $2.50 output per million tokens (median across 14 API providers). At a typical mix of three input tokens to one output token, that is $0.493 per million tokens for WizardLM 2 8x22B versus $0.72 for Llama-3.1-70B-Instruct (1.5× as much) and $1.07 for Kimi K2 Thinking (2.2× as much).
Which scores higher on benchmarks?
Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond): Kimi K2 Thinking 84.2%, Llama-3.1-70B-Instruct 44.2% and WizardLM 2 8x22B 43.4%. On individual benchmarks: GPQA Diamond — Kimi K2 Thinking 84.2%, Llama-3.1-70B-Instruct 44.2%, WizardLM 2 8x22B 43.4%.
Which is better for coding?
There are no published SWE-bench Verified results for Kimi K2 Thinking, Llama-3.1-70B-Instruct and WizardLM 2 8x22B yet, so there is no like-for-like coding score. On overall capability, Kimi K2 Thinking leads, which tends to carry over to coding, but test on your own codebase. Note that WizardLM 2 8x22B does not support tool calling, which most coding agents need.
Which has the bigger context window?
Kimi K2 Thinking has the largest context window at 262,144 tokens, against 128,000 for Llama-3.1-70B-Instruct and 65,536 for WizardLM 2 8x22B. Maximum output per response: Kimi K2 Thinking up to 262,144, Llama-3.1-70B-Instruct up to 4,096, WizardLM 2 8x22B up to 8,000 tokens.
Which can read images, PDFs, audio or video?
Kimi K2 Thinking accepts text; Llama-3.1-70B-Instruct accepts text; WizardLM 2 8x22B accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights (Apache-2.0), so you can self-host them.
Which is newer?
Kimi K2 Thinking is the newest, released Nov 6, 2025. Llama-3.1-70B-Instruct came out Jul 23, 2024; WizardLM 2 8x22B came out Apr 15, 2024. Knowledge cutoff: Kimi K2 Thinking Aug 2024, Llama-3.1-70B-Instruct Dec 2023.
How do you decide the winner?
This rating uses 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 Nov 18, 2026.